feat: issue_close + issue_reopen tools (#30)

Adds two MCP tools that PATCH /api/v1/repos/{owner}/{name}/issues/{number}
with {"state":"closed"} or {"state":"open"}. Both use a shared
SetIssueState helper on the gitea client.

- internal/gitea/issues.go: SetIssueState method using the existing
  PatchJSON + MapStatus + json.Unmarshal pattern from GetIssue.
- internal/tools/issue_close.go: IssueClose tool. owner+name+number
  args. Owner allowlist enforced. Returns the updated issue. Reversible
  via issue_reopen, classified LOW risk.
- internal/tools/issue_reopen.go: mirror of IssueClose with
  state="open". Same risk profile.
- Registered both tools in cmd/gitea-mcp/main.go.
- Tests for both: success (asserts PATCH method, path, body), 404,
  and allowlist rejection — same shape as issue_get_test.go.

Closes #30

.aider.conventions.md

@@ -36,9 +36,18 @@ These rules apply to every task across every project, regardless of harness.
 4. **Goal-driven execution.** Define clear success criteria up front for every task.
    Loop — implement, verify, refine — until those criteria are met. Don't claim
    completion without evidence (tests pass, command output, observed behavior).
-5. **Branch-per-task for multi-agent repos.** When another agent may be active on
-   the same repo, create a branch (`agent/<description>`), commit there, and open a
-   PR. Do not merge without explicit instruction from Mathias.
+5. **Trunk-Based Development — commit directly to main.** Every commit is one
+   logical change (one tool, one fix, one test) with passing tests. Main is always
+   deployable. Never create long-lived feature branches.
+
+   **Exception — parallel agents on same repo:** If another agent is known to be
+   actively working on the same repo simultaneously, create a short-lived branch
+   (`agent/<description>`), finish the task, and merge to main within the same
+   session. Do not leave agent branches open between sessions.
+
+   **Exception — external contributor or client four-eyes requirement:** Use
+   PR flow only when a human reviewer outside the project is required. Document
+   the reason in PROJECT.md.
 
 ## Default stack
 
@@ -62,7 +71,10 @@ Exploratory: Rust, Zig — I'll tell you when I want these.
 - **Errors**: `fmt.Errorf("operation: %w", err)` — never naked, never log-and-return
 - **Naming**: stdlib conventions, no stuttering
 - **Architecture**: prefer stdlib over frameworks, constructor injection, env-var config parsed into typed structs
-- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), one concern per PR, PR describes *why* not *what*
+- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), commit directly to main,
+  one logical change per commit, CI is the quality gate
+- **Never**: long-lived feature branches, PRs for solo work, direct push without
+  passing `task check` locally first
 - **Security**: no secrets in code, govulncheck before adding deps, SOPS for encrypted config
 - **Dependencies**: prefer stdlib. testify, slog, templ, sqlc, google.golang.org/adk (agent projects only) are pre-approved; anything else needs justification in the commit message
 
@@ -104,18 +116,64 @@ See `~/dev/PROJECT_SUMMARY.md` for detailed descriptions of each project.
 - **koala-ai-stack** (`AGENTS/`) — local AI server infrastructure management
 - **klimatkollen** (`XT/`) — Swedish municipal climate data platform
 
-## Knowledge base
+## Knowledge base — actively use it
 
-When available, agents can query the shared knowledge base:
+A persistent brain (BM25 search + LLM-synthesised Q&A) survives across sessions,
+hosts, and harnesses. It holds 100+ hard-won entries: infra incident postmortems,
+Go pitfalls, framework gotchas, design principles, ADRs. **It is not optional
+reference material — query it actively, not just when explicitly told.**
 
-- **MCP**: `mcp://hyperguild.<TAILNET>.ts.net:3100/knowledge`
-- **HTTP**: `http://hyperguild.<TAILNET>.ts.net:3100/api/v1/search`
+### When to query (treat as a reflex)
 
-<!-- TODO: replace <TAILNET> placeholder with the real Tailscale tailnet
-     name once hyperguild is deployed. Until then, agents that try to
-     reach the knowledge service on a host where it isn't running will
-     get DNS NXDOMAIN, which is the desired fail-loudly behavior. -->
-- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`
+- **Before** starting a non-trivial task — search for prior art with the symptom
+  AND the system component ("how did we solve X in Y?"). 5 seconds beats 5 hours.
+- **When debugging** — search for the error string, the stack frame, the affected
+  service. Past you may have already paid this tax.
+- **Before adopting** a pattern, library, framework, or model name — check if it
+  was tried and rejected, or what the integration footguns are.
+- **When making architectural decisions** — search for the domain + "ADR" or
+  "decision" to find prior reasoning before re-deriving it.
+- **When a recommendation feels novel** — challenge yourself: "has this been
+  documented?" The brain often has it.
+
+### When to write
+
+After you discover something that **future-you would forget** and that **isn't
+recoverable from the code, git log, or PR description alone**:
+
+- Bugs whose root cause is non-obvious and generalisable beyond this project.
+- Framework / library / model-name quirks that bit you and would bite anyone.
+- Design principles validated under fire (e.g. "every `_get` needs a `_list`").
+- Postmortems for incidents: what broke, why, how diagnosed, what to do next time.
+
+DON'T write project status, sprint progress, PR summaries, or "what I did this
+session" — those rot fast and the originals are in git/gitea anyway. Brain
+entries that age well are about *why*, *how to avoid*, and *what to do when*.
+
+### How to access (per harness)
+
+| Harness | Query | Write |
+|---------|-------|-------|
+| **Claude Code, Claude Desktop** | `brain_query` (BM25), `brain_answer` (LLM-synth + sources) MCP tools | `brain_write` MCP tool |
+| **Crush, Pi, Antigravity, other MCP-capable** | same MCP server: `ingestion-brain` (via the `mcp__*_brain__*` namespace once authenticated) | same |
+| **Anything HTTP-only (curl, scripts)** | `POST https://brain-mcp.d-ma.be/query` with `{"query":"..."}` (auth via `BRAIN_MCP_TOKEN`) | `POST .../write` with `{"content":"...","filename":"..."}` |
+| **Browser / human inspection** | `https://gitea.d-ma.be/mathias/hyperguild` → `knowledge/` and `wiki/` markdown files |
+
+- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`.
+- **Routing**: brain_answer's LLM uses berget.ai as primary, iguana ollama as
+  fallback. Both are configurable in the `supervisor/ingestion-deployment.yaml`
+  on the koala k3s cluster; don't hardcode local-only model names into the
+  berget URL (see knowledge entry on namespace mismatches).
+
+### Quick reflex checks
+
+If you find yourself about to say any of these out loud, you owe yourself a brain query first:
+
+- "I think the issue might be..."
+- "Let me try X and see..."
+- "I'll just write a script to..."
+- "This is probably a new bug..."
+- "Has anyone done this before?" — *yes, probably, go check.*
 
 ## Client work rules
 
@@ -212,8 +270,11 @@ Key skills:
 
 ### Git
 - Conventional commits: `feat:`, `fix:`, `chore:`, `docs:`, `refactor:`
-- Branch naming: `feat/short-description`, `fix/short-description`
-- PRs: one concern per PR, description explains *why* not *what*
+- **Trunk-Based Development:** commit directly to main. One logical change per commit.
+- Run `task check` locally before every push. CI is the quality gate, not branch protection.
+- No feature branches, no PRs for solo/agent work.
+- Exception: if a parallel agent session is active on this repo, use a short-lived
+  `agent/<description>` branch and merge within the same session.
 
 ### Security
 - No secrets in code, ever — use env vars or SOPS-encrypted files
@@ -251,68 +312,29 @@ When acting as a coding agent on this project:
 3. If unsure about a convention, check `DECISIONS.md` or ask
 4. Never modify files outside the project root without explicit permission
 5. When adding a dependency, explain why in the commit message
-6. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
+6. Commit directly to main. Run `task check` before every push. Never create
+   feature branches unless a parallel agent is simultaneously active on this repo.
+7. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
 
-## Current sprint — gitea-mcp v0.2 (2026-05-14)
+## Current state — v0.2.5 (2026-05-17)
 
-### Context
-This sprint implements new MCP tools needed for `hyperguild new-project` —
-the automated project creation flow triggered from claude.ai. See brain knowledge
-nodes `adr-new-project-gitea-first-github-mirror` and `roadmap-github-ingestion-pipeline`
-for full background.
+All v0.2 work is complete and deployed. No active sprint.
 
-### Issues to implement (priority order)
+### What shipped
 
-**Batch 1 — blockers (do first, one PR: `feat/repo-crud`)**
+| Tag | PR | Tools / fixes |
+|-----|----|---------------|
+| v0.2.2 | #21 | repo_create, repo_update, repo_mirror_push |
+| v0.2.3 | #22 | repo_tree, repo_topics_update, file_read dir fix |
+| v0.2.4 | #23 | issue_get, release_create, repo_delete |
+| v0.2.5 | #26 | repo_update archived+template, create_project_from_template template_name, pr_files_diff loop fix |
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #13 | `repo_create` | POST /api/v1/user/repos or /api/v1/orgs/{org}/repos |
-| #16 | `repo_mirror_push` (add/list/delete) | POST/GET/DELETE /api/v1/repos/{owner}/{repo}/push_mirrors |
-| #12 | `repo_update` | PATCH /api/v1/repos/{owner}/{repo} |
+Current main: `e31fd3f`. CI green. Deployed via Flux.
 
-**Batch 2 — quality of life (second PR: `feat/repo-ux`)**
+### Next up
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #15 | `file_read` dir-path fix | existing endpoint, detect array vs object response |
-| #14 | `repo_tree` | GET /api/v1/repos/{owner}/{repo}/git/trees/{sha}?recursive=true |
-| #18 | `repo_topics_update` | PUT /api/v1/repos/{owner}/{repo}/topics |
+1. **`hyperguild new-project` v1** — primary next target.
+   See brain node `adr-new-project-gitea-first-github-mirror` for full flow spec.
 
-**Batch 3 — can wait**
-
-| Issue | Tool | Note |
-|-------|------|------|
-| #11 | `repo_delete` | HIGH risk — needs `confirm` param == repo name |
-| #17 | `release_create` | POST /api/v1/repos/{owner}/{repo}/releases |
-
-### How to add a tool (pattern)
-
-Every tool = 4 files following `internal/tools/repo_get.go` exactly:
-
-1. `internal/gitea/<domain>.go` — API client method (use PostJSON/PatchJSON/DeleteJSON)
-2. `internal/tools/repo_<name>.go` — tool handler with Descriptor() + Call()
-3. `internal/tools/repo_<name>_test.go` — table-driven tests with httptest.NewServer
-4. Registration in main — find where `NewRepoGet` is registered, add new tool same place
-
-Key rules:
-- Always call `t.a.Check(args.Owner)` before any API call (allowlist guard)
-- Use `textOK(result)` for success output
-- For `repo_mirror_push`: NEVER log or return `remote_password` in any output
-- For `repo_update` with `private: false` and `repo_delete`: require `confirm` param == repo name
-
-### Token permissions needed
-
-New tools require these additional Gitea token scopes:
-- `write:repository` — repo_create, repo_update, repo_mirror_push, repo_topics_update, release_create
-- `delete_repo` — repo_delete
-
-Check current token: `curl -H "Authorization: token $GITEA_TOKEN" https://gitea.d-ma.be/api/v1/user`
-If scopes are missing, update token in Gitea settings before running tests.
-
-### Definition of done
-
-- `task check` passes (all tools, all batches)
-- Each new tool manually callable via `claude mcp call`
-- PR #1 (batch 1) merged before starting batch 2
-- Issue #19 (mirror flow e2e test) verified manually after batch 1 is deployed
+2. **Issue #19** — end-to-end mirror flow verification.
+   `repo_mirror_push` is implemented but the full flow (create repo → add push mirror → verify sync to GitHub) has not been tested manually. Do this before relying on it in production.

.context/PROJECT.md

@@ -37,8 +37,11 @@
 
 ### Git
 - Conventional commits: `feat:`, `fix:`, `chore:`, `docs:`, `refactor:`
-- Branch naming: `feat/short-description`, `fix/short-description`
-- PRs: one concern per PR, description explains *why* not *what*
+- **Trunk-Based Development:** commit directly to main. One logical change per commit.
+- Run `task check` locally before every push. CI is the quality gate, not branch protection.
+- No feature branches, no PRs for solo/agent work.
+- Exception: if a parallel agent session is active on this repo, use a short-lived
+  `agent/<description>` branch and merge within the same session.
 
 ### Security
 - No secrets in code, ever — use env vars or SOPS-encrypted files
@@ -76,68 +79,29 @@ When acting as a coding agent on this project:
 3. If unsure about a convention, check `DECISIONS.md` or ask
 4. Never modify files outside the project root without explicit permission
 5. When adding a dependency, explain why in the commit message
-6. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
+6. Commit directly to main. Run `task check` before every push. Never create
+   feature branches unless a parallel agent is simultaneously active on this repo.
+7. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
 
-## Current sprint — gitea-mcp v0.2 (2026-05-14)
+## Current state — v0.2.5 (2026-05-17)
 
-### Context
-This sprint implements new MCP tools needed for `hyperguild new-project` —
-the automated project creation flow triggered from claude.ai. See brain knowledge
-nodes `adr-new-project-gitea-first-github-mirror` and `roadmap-github-ingestion-pipeline`
-for full background.
+All v0.2 work is complete and deployed. No active sprint.
 
-### Issues to implement (priority order)
+### What shipped
 
-**Batch 1 — blockers (do first, one PR: `feat/repo-crud`)**
+| Tag | PR | Tools / fixes |
+|-----|----|---------------|
+| v0.2.2 | #21 | repo_create, repo_update, repo_mirror_push |
+| v0.2.3 | #22 | repo_tree, repo_topics_update, file_read dir fix |
+| v0.2.4 | #23 | issue_get, release_create, repo_delete |
+| v0.2.5 | #26 | repo_update archived+template, create_project_from_template template_name, pr_files_diff loop fix |
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #13 | `repo_create` | POST /api/v1/user/repos or /api/v1/orgs/{org}/repos |
-| #16 | `repo_mirror_push` (add/list/delete) | POST/GET/DELETE /api/v1/repos/{owner}/{repo}/push_mirrors |
-| #12 | `repo_update` | PATCH /api/v1/repos/{owner}/{repo} |
+Current main: `e31fd3f`. CI green. Deployed via Flux.
 
-**Batch 2 — quality of life (second PR: `feat/repo-ux`)**
+### Next up
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #15 | `file_read` dir-path fix | existing endpoint, detect array vs object response |
-| #14 | `repo_tree` | GET /api/v1/repos/{owner}/{repo}/git/trees/{sha}?recursive=true |
-| #18 | `repo_topics_update` | PUT /api/v1/repos/{owner}/{repo}/topics |
+1. **`hyperguild new-project` v1** — primary next target.
+   See brain node `adr-new-project-gitea-first-github-mirror` for full flow spec.
 
-**Batch 3 — can wait**
-
-| Issue | Tool | Note |
-|-------|------|------|
-| #11 | `repo_delete` | HIGH risk — needs `confirm` param == repo name |
-| #17 | `release_create` | POST /api/v1/repos/{owner}/{repo}/releases |
-
-### How to add a tool (pattern)
-
-Every tool = 4 files following `internal/tools/repo_get.go` exactly:
-
-1. `internal/gitea/<domain>.go` — API client method (use PostJSON/PatchJSON/DeleteJSON)
-2. `internal/tools/repo_<name>.go` — tool handler with Descriptor() + Call()
-3. `internal/tools/repo_<name>_test.go` — table-driven tests with httptest.NewServer
-4. Registration in main — find where `NewRepoGet` is registered, add new tool same place
-
-Key rules:
-- Always call `t.a.Check(args.Owner)` before any API call (allowlist guard)
-- Use `textOK(result)` for success output
-- For `repo_mirror_push`: NEVER log or return `remote_password` in any output
-- For `repo_update` with `private: false` and `repo_delete`: require `confirm` param == repo name
-
-### Token permissions needed
-
-New tools require these additional Gitea token scopes:
-- `write:repository` — repo_create, repo_update, repo_mirror_push, repo_topics_update, release_create
-- `delete_repo` — repo_delete
-
-Check current token: `curl -H "Authorization: token $GITEA_TOKEN" https://gitea.d-ma.be/api/v1/user`
-If scopes are missing, update token in Gitea settings before running tests.
-
-### Definition of done
-
-- `task check` passes (all tools, all batches)
-- Each new tool manually callable via `claude mcp call`
-- PR #1 (batch 1) merged before starting batch 2
-- Issue #19 (mirror flow e2e test) verified manually after batch 1 is deployed
+2. **Issue #19** — end-to-end mirror flow verification.
+   `repo_mirror_push` is implemented but the full flow (create repo → add push mirror → verify sync to GitHub) has not been tested manually. Do this before relying on it in production.

.context/system-prompt.txt

@@ -41,9 +41,18 @@ These rules apply to every task across every project, regardless of harness.
 4. **Goal-driven execution.** Define clear success criteria up front for every task.
    Loop — implement, verify, refine — until those criteria are met. Don't claim
    completion without evidence (tests pass, command output, observed behavior).
-5. **Branch-per-task for multi-agent repos.** When another agent may be active on
-   the same repo, create a branch (`agent/<description>`), commit there, and open a
-   PR. Do not merge without explicit instruction from Mathias.
+5. **Trunk-Based Development — commit directly to main.** Every commit is one
+   logical change (one tool, one fix, one test) with passing tests. Main is always
+   deployable. Never create long-lived feature branches.
+
+   **Exception — parallel agents on same repo:** If another agent is known to be
+   actively working on the same repo simultaneously, create a short-lived branch
+   (`agent/<description>`), finish the task, and merge to main within the same
+   session. Do not leave agent branches open between sessions.
+
+   **Exception — external contributor or client four-eyes requirement:** Use
+   PR flow only when a human reviewer outside the project is required. Document
+   the reason in PROJECT.md.
 
 ## Default stack
 
@@ -67,7 +76,10 @@ Exploratory: Rust, Zig — I'll tell you when I want these.
 - **Errors**: `fmt.Errorf("operation: %w", err)` — never naked, never log-and-return
 - **Naming**: stdlib conventions, no stuttering
 - **Architecture**: prefer stdlib over frameworks, constructor injection, env-var config parsed into typed structs
-- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), one concern per PR, PR describes *why* not *what*
+- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), commit directly to main,
+  one logical change per commit, CI is the quality gate
+- **Never**: long-lived feature branches, PRs for solo work, direct push without
+  passing `task check` locally first
 - **Security**: no secrets in code, govulncheck before adding deps, SOPS for encrypted config
 - **Dependencies**: prefer stdlib. testify, slog, templ, sqlc, google.golang.org/adk (agent projects only) are pre-approved; anything else needs justification in the commit message
 
@@ -109,18 +121,64 @@ See `~/dev/PROJECT_SUMMARY.md` for detailed descriptions of each project.
 - **koala-ai-stack** (`AGENTS/`) — local AI server infrastructure management
 - **klimatkollen** (`XT/`) — Swedish municipal climate data platform
 
-## Knowledge base
+## Knowledge base — actively use it
 
-When available, agents can query the shared knowledge base:
+A persistent brain (BM25 search + LLM-synthesised Q&A) survives across sessions,
+hosts, and harnesses. It holds 100+ hard-won entries: infra incident postmortems,
+Go pitfalls, framework gotchas, design principles, ADRs. **It is not optional
+reference material — query it actively, not just when explicitly told.**
 
-- **MCP**: `mcp://hyperguild.<TAILNET>.ts.net:3100/knowledge`
-- **HTTP**: `http://hyperguild.<TAILNET>.ts.net:3100/api/v1/search`
+### When to query (treat as a reflex)
 
-<!-- TODO: replace <TAILNET> placeholder with the real Tailscale tailnet
-     name once hyperguild is deployed. Until then, agents that try to
-     reach the knowledge service on a host where it isn't running will
-     get DNS NXDOMAIN, which is the desired fail-loudly behavior. -->
-- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`
+- **Before** starting a non-trivial task — search for prior art with the symptom
+  AND the system component ("how did we solve X in Y?"). 5 seconds beats 5 hours.
+- **When debugging** — search for the error string, the stack frame, the affected
+  service. Past you may have already paid this tax.
+- **Before adopting** a pattern, library, framework, or model name — check if it
+  was tried and rejected, or what the integration footguns are.
+- **When making architectural decisions** — search for the domain + "ADR" or
+  "decision" to find prior reasoning before re-deriving it.
+- **When a recommendation feels novel** — challenge yourself: "has this been
+  documented?" The brain often has it.
+
+### When to write
+
+After you discover something that **future-you would forget** and that **isn't
+recoverable from the code, git log, or PR description alone**:
+
+- Bugs whose root cause is non-obvious and generalisable beyond this project.
+- Framework / library / model-name quirks that bit you and would bite anyone.
+- Design principles validated under fire (e.g. "every `_get` needs a `_list`").
+- Postmortems for incidents: what broke, why, how diagnosed, what to do next time.
+
+DON'T write project status, sprint progress, PR summaries, or "what I did this
+session" — those rot fast and the originals are in git/gitea anyway. Brain
+entries that age well are about *why*, *how to avoid*, and *what to do when*.
+
+### How to access (per harness)
+
+| Harness | Query | Write |
+|---------|-------|-------|
+| **Claude Code, Claude Desktop** | `brain_query` (BM25), `brain_answer` (LLM-synth + sources) MCP tools | `brain_write` MCP tool |
+| **Crush, Pi, Antigravity, other MCP-capable** | same MCP server: `ingestion-brain` (via the `mcp__*_brain__*` namespace once authenticated) | same |
+| **Anything HTTP-only (curl, scripts)** | `POST https://brain-mcp.d-ma.be/query` with `{"query":"..."}` (auth via `BRAIN_MCP_TOKEN`) | `POST .../write` with `{"content":"...","filename":"..."}` |
+| **Browser / human inspection** | `https://gitea.d-ma.be/mathias/hyperguild` → `knowledge/` and `wiki/` markdown files |
+
+- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`.
+- **Routing**: brain_answer's LLM uses berget.ai as primary, iguana ollama as
+  fallback. Both are configurable in the `supervisor/ingestion-deployment.yaml`
+  on the koala k3s cluster; don't hardcode local-only model names into the
+  berget URL (see knowledge entry on namespace mismatches).
+
+### Quick reflex checks
+
+If you find yourself about to say any of these out loud, you owe yourself a brain query first:
+
+- "I think the issue might be..."
+- "Let me try X and see..."
+- "I'll just write a script to..."
+- "This is probably a new bug..."
+- "Has anyone done this before?" — *yes, probably, go check.*
 
 ## Client work rules
 
@@ -217,8 +275,11 @@ Key skills:
 
 ### Git
 - Conventional commits: `feat:`, `fix:`, `chore:`, `docs:`, `refactor:`
-- Branch naming: `feat/short-description`, `fix/short-description`
-- PRs: one concern per PR, description explains *why* not *what*
+- **Trunk-Based Development:** commit directly to main. One logical change per commit.
+- Run `task check` locally before every push. CI is the quality gate, not branch protection.
+- No feature branches, no PRs for solo/agent work.
+- Exception: if a parallel agent session is active on this repo, use a short-lived
+  `agent/<description>` branch and merge within the same session.
 
 ### Security
 - No secrets in code, ever — use env vars or SOPS-encrypted files
@@ -256,70 +317,31 @@ When acting as a coding agent on this project:
 3. If unsure about a convention, check `DECISIONS.md` or ask
 4. Never modify files outside the project root without explicit permission
 5. When adding a dependency, explain why in the commit message
-6. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
+6. Commit directly to main. Run `task check` before every push. Never create
+   feature branches unless a parallel agent is simultaneously active on this repo.
+7. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
 
-## Current sprint — gitea-mcp v0.2 (2026-05-14)
+## Current state — v0.2.5 (2026-05-17)
 
-### Context
-This sprint implements new MCP tools needed for `hyperguild new-project` —
-the automated project creation flow triggered from claude.ai. See brain knowledge
-nodes `adr-new-project-gitea-first-github-mirror` and `roadmap-github-ingestion-pipeline`
-for full background.
+All v0.2 work is complete and deployed. No active sprint.
 
-### Issues to implement (priority order)
+### What shipped
 
-**Batch 1 — blockers (do first, one PR: `feat/repo-crud`)**
+| Tag | PR | Tools / fixes |
+|-----|----|---------------|
+| v0.2.2 | #21 | repo_create, repo_update, repo_mirror_push |
+| v0.2.3 | #22 | repo_tree, repo_topics_update, file_read dir fix |
+| v0.2.4 | #23 | issue_get, release_create, repo_delete |
+| v0.2.5 | #26 | repo_update archived+template, create_project_from_template template_name, pr_files_diff loop fix |
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #13 | `repo_create` | POST /api/v1/user/repos or /api/v1/orgs/{org}/repos |
-| #16 | `repo_mirror_push` (add/list/delete) | POST/GET/DELETE /api/v1/repos/{owner}/{repo}/push_mirrors |
-| #12 | `repo_update` | PATCH /api/v1/repos/{owner}/{repo} |
+Current main: `e31fd3f`. CI green. Deployed via Flux.
 
-**Batch 2 — quality of life (second PR: `feat/repo-ux`)**
+### Next up
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #15 | `file_read` dir-path fix | existing endpoint, detect array vs object response |
-| #14 | `repo_tree` | GET /api/v1/repos/{owner}/{repo}/git/trees/{sha}?recursive=true |
-| #18 | `repo_topics_update` | PUT /api/v1/repos/{owner}/{repo}/topics |
+1. **`hyperguild new-project` v1** — primary next target.
+   See brain node `adr-new-project-gitea-first-github-mirror` for full flow spec.
 
-**Batch 3 — can wait**
-
-| Issue | Tool | Note |
-|-------|------|------|
-| #11 | `repo_delete` | HIGH risk — needs `confirm` param == repo name |
-| #17 | `release_create` | POST /api/v1/repos/{owner}/{repo}/releases |
-
-### How to add a tool (pattern)
-
-Every tool = 4 files following `internal/tools/repo_get.go` exactly:
-
-1. `internal/gitea/<domain>.go` — API client method (use PostJSON/PatchJSON/DeleteJSON)
-2. `internal/tools/repo_<name>.go` — tool handler with Descriptor() + Call()
-3. `internal/tools/repo_<name>_test.go` — table-driven tests with httptest.NewServer
-4. Registration in main — find where `NewRepoGet` is registered, add new tool same place
-
-Key rules:
-- Always call `t.a.Check(args.Owner)` before any API call (allowlist guard)
-- Use `textOK(result)` for success output
-- For `repo_mirror_push`: NEVER log or return `remote_password` in any output
-- For `repo_update` with `private: false` and `repo_delete`: require `confirm` param == repo name
-
-### Token permissions needed
-
-New tools require these additional Gitea token scopes:
-- `write:repository` — repo_create, repo_update, repo_mirror_push, repo_topics_update, release_create
-- `delete_repo` — repo_delete
-
-Check current token: `curl -H "Authorization: token $GITEA_TOKEN" https://gitea.d-ma.be/api/v1/user`
-If scopes are missing, update token in Gitea settings before running tests.
-
-### Definition of done
-
-- `task check` passes (all tools, all batches)
-- Each new tool manually callable via `claude mcp call`
-- PR #1 (batch 1) merged before starting batch 2
-- Issue #19 (mirror flow e2e test) verified manually after batch 1 is deployed
+2. **Issue #19** — end-to-end mirror flow verification.
+   `repo_mirror_push` is implemented but the full flow (create repo → add push mirror → verify sync to GitHub) has not been tested manually. Do this before relying on it in production.
 
 ---

.cursorrules

@@ -39,9 +39,18 @@ These rules apply to every task across every project, regardless of harness.
 4. **Goal-driven execution.** Define clear success criteria up front for every task.
    Loop — implement, verify, refine — until those criteria are met. Don't claim
    completion without evidence (tests pass, command output, observed behavior).
-5. **Branch-per-task for multi-agent repos.** When another agent may be active on
-   the same repo, create a branch (`agent/<description>`), commit there, and open a
-   PR. Do not merge without explicit instruction from Mathias.
+5. **Trunk-Based Development — commit directly to main.** Every commit is one
+   logical change (one tool, one fix, one test) with passing tests. Main is always
+   deployable. Never create long-lived feature branches.
+
+   **Exception — parallel agents on same repo:** If another agent is known to be
+   actively working on the same repo simultaneously, create a short-lived branch
+   (`agent/<description>`), finish the task, and merge to main within the same
+   session. Do not leave agent branches open between sessions.
+
+   **Exception — external contributor or client four-eyes requirement:** Use
+   PR flow only when a human reviewer outside the project is required. Document
+   the reason in PROJECT.md.
 
 ## Default stack
 
@@ -65,7 +74,10 @@ Exploratory: Rust, Zig — I'll tell you when I want these.
 - **Errors**: `fmt.Errorf("operation: %w", err)` — never naked, never log-and-return
 - **Naming**: stdlib conventions, no stuttering
 - **Architecture**: prefer stdlib over frameworks, constructor injection, env-var config parsed into typed structs
-- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), one concern per PR, PR describes *why* not *what*
+- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), commit directly to main,
+  one logical change per commit, CI is the quality gate
+- **Never**: long-lived feature branches, PRs for solo work, direct push without
+  passing `task check` locally first
 - **Security**: no secrets in code, govulncheck before adding deps, SOPS for encrypted config
 - **Dependencies**: prefer stdlib. testify, slog, templ, sqlc, google.golang.org/adk (agent projects only) are pre-approved; anything else needs justification in the commit message
 
@@ -107,18 +119,64 @@ See `~/dev/PROJECT_SUMMARY.md` for detailed descriptions of each project.
 - **koala-ai-stack** (`AGENTS/`) — local AI server infrastructure management
 - **klimatkollen** (`XT/`) — Swedish municipal climate data platform
 
-## Knowledge base
+## Knowledge base — actively use it
 
-When available, agents can query the shared knowledge base:
+A persistent brain (BM25 search + LLM-synthesised Q&A) survives across sessions,
+hosts, and harnesses. It holds 100+ hard-won entries: infra incident postmortems,
+Go pitfalls, framework gotchas, design principles, ADRs. **It is not optional
+reference material — query it actively, not just when explicitly told.**
 
-- **MCP**: `mcp://hyperguild.<TAILNET>.ts.net:3100/knowledge`
-- **HTTP**: `http://hyperguild.<TAILNET>.ts.net:3100/api/v1/search`
+### When to query (treat as a reflex)
 
-<!-- TODO: replace <TAILNET> placeholder with the real Tailscale tailnet
-     name once hyperguild is deployed. Until then, agents that try to
-     reach the knowledge service on a host where it isn't running will
-     get DNS NXDOMAIN, which is the desired fail-loudly behavior. -->
-- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`
+- **Before** starting a non-trivial task — search for prior art with the symptom
+  AND the system component ("how did we solve X in Y?"). 5 seconds beats 5 hours.
+- **When debugging** — search for the error string, the stack frame, the affected
+  service. Past you may have already paid this tax.
+- **Before adopting** a pattern, library, framework, or model name — check if it
+  was tried and rejected, or what the integration footguns are.
+- **When making architectural decisions** — search for the domain + "ADR" or
+  "decision" to find prior reasoning before re-deriving it.
+- **When a recommendation feels novel** — challenge yourself: "has this been
+  documented?" The brain often has it.
+
+### When to write
+
+After you discover something that **future-you would forget** and that **isn't
+recoverable from the code, git log, or PR description alone**:
+
+- Bugs whose root cause is non-obvious and generalisable beyond this project.
+- Framework / library / model-name quirks that bit you and would bite anyone.
+- Design principles validated under fire (e.g. "every `_get` needs a `_list`").
+- Postmortems for incidents: what broke, why, how diagnosed, what to do next time.
+
+DON'T write project status, sprint progress, PR summaries, or "what I did this
+session" — those rot fast and the originals are in git/gitea anyway. Brain
+entries that age well are about *why*, *how to avoid*, and *what to do when*.
+
+### How to access (per harness)
+
+| Harness | Query | Write |
+|---------|-------|-------|
+| **Claude Code, Claude Desktop** | `brain_query` (BM25), `brain_answer` (LLM-synth + sources) MCP tools | `brain_write` MCP tool |
+| **Crush, Pi, Antigravity, other MCP-capable** | same MCP server: `ingestion-brain` (via the `mcp__*_brain__*` namespace once authenticated) | same |
+| **Anything HTTP-only (curl, scripts)** | `POST https://brain-mcp.d-ma.be/query` with `{"query":"..."}` (auth via `BRAIN_MCP_TOKEN`) | `POST .../write` with `{"content":"...","filename":"..."}` |
+| **Browser / human inspection** | `https://gitea.d-ma.be/mathias/hyperguild` → `knowledge/` and `wiki/` markdown files |
+
+- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`.
+- **Routing**: brain_answer's LLM uses berget.ai as primary, iguana ollama as
+  fallback. Both are configurable in the `supervisor/ingestion-deployment.yaml`
+  on the koala k3s cluster; don't hardcode local-only model names into the
+  berget URL (see knowledge entry on namespace mismatches).
+
+### Quick reflex checks
+
+If you find yourself about to say any of these out loud, you owe yourself a brain query first:
+
+- "I think the issue might be..."
+- "Let me try X and see..."
+- "I'll just write a script to..."
+- "This is probably a new bug..."
+- "Has anyone done this before?" — *yes, probably, go check.*
 
 ## Client work rules
 
@@ -215,8 +273,11 @@ Key skills:
 
 ### Git
 - Conventional commits: `feat:`, `fix:`, `chore:`, `docs:`, `refactor:`
-- Branch naming: `feat/short-description`, `fix/short-description`
-- PRs: one concern per PR, description explains *why* not *what*
+- **Trunk-Based Development:** commit directly to main. One logical change per commit.
+- Run `task check` locally before every push. CI is the quality gate, not branch protection.
+- No feature branches, no PRs for solo/agent work.
+- Exception: if a parallel agent session is active on this repo, use a short-lived
+  `agent/<description>` branch and merge within the same session.
 
 ### Security
 - No secrets in code, ever — use env vars or SOPS-encrypted files
@@ -254,68 +315,29 @@ When acting as a coding agent on this project:
 3. If unsure about a convention, check `DECISIONS.md` or ask
 4. Never modify files outside the project root without explicit permission
 5. When adding a dependency, explain why in the commit message
-6. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
+6. Commit directly to main. Run `task check` before every push. Never create
+   feature branches unless a parallel agent is simultaneously active on this repo.
+7. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
 
-## Current sprint — gitea-mcp v0.2 (2026-05-14)
+## Current state — v0.2.5 (2026-05-17)
 
-### Context
-This sprint implements new MCP tools needed for `hyperguild new-project` —
-the automated project creation flow triggered from claude.ai. See brain knowledge
-nodes `adr-new-project-gitea-first-github-mirror` and `roadmap-github-ingestion-pipeline`
-for full background.
+All v0.2 work is complete and deployed. No active sprint.
 
-### Issues to implement (priority order)
+### What shipped
 
-**Batch 1 — blockers (do first, one PR: `feat/repo-crud`)**
+| Tag | PR | Tools / fixes |
+|-----|----|---------------|
+| v0.2.2 | #21 | repo_create, repo_update, repo_mirror_push |
+| v0.2.3 | #22 | repo_tree, repo_topics_update, file_read dir fix |
+| v0.2.4 | #23 | issue_get, release_create, repo_delete |
+| v0.2.5 | #26 | repo_update archived+template, create_project_from_template template_name, pr_files_diff loop fix |
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #13 | `repo_create` | POST /api/v1/user/repos or /api/v1/orgs/{org}/repos |
-| #16 | `repo_mirror_push` (add/list/delete) | POST/GET/DELETE /api/v1/repos/{owner}/{repo}/push_mirrors |
-| #12 | `repo_update` | PATCH /api/v1/repos/{owner}/{repo} |
+Current main: `e31fd3f`. CI green. Deployed via Flux.
 
-**Batch 2 — quality of life (second PR: `feat/repo-ux`)**
+### Next up
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #15 | `file_read` dir-path fix | existing endpoint, detect array vs object response |
-| #14 | `repo_tree` | GET /api/v1/repos/{owner}/{repo}/git/trees/{sha}?recursive=true |
-| #18 | `repo_topics_update` | PUT /api/v1/repos/{owner}/{repo}/topics |
+1. **`hyperguild new-project` v1** — primary next target.
+   See brain node `adr-new-project-gitea-first-github-mirror` for full flow spec.
 
-**Batch 3 — can wait**
-
-| Issue | Tool | Note |
-|-------|------|------|
-| #11 | `repo_delete` | HIGH risk — needs `confirm` param == repo name |
-| #17 | `release_create` | POST /api/v1/repos/{owner}/{repo}/releases |
-
-### How to add a tool (pattern)
-
-Every tool = 4 files following `internal/tools/repo_get.go` exactly:
-
-1. `internal/gitea/<domain>.go` — API client method (use PostJSON/PatchJSON/DeleteJSON)
-2. `internal/tools/repo_<name>.go` — tool handler with Descriptor() + Call()
-3. `internal/tools/repo_<name>_test.go` — table-driven tests with httptest.NewServer
-4. Registration in main — find where `NewRepoGet` is registered, add new tool same place
-
-Key rules:
-- Always call `t.a.Check(args.Owner)` before any API call (allowlist guard)
-- Use `textOK(result)` for success output
-- For `repo_mirror_push`: NEVER log or return `remote_password` in any output
-- For `repo_update` with `private: false` and `repo_delete`: require `confirm` param == repo name
-
-### Token permissions needed
-
-New tools require these additional Gitea token scopes:
-- `write:repository` — repo_create, repo_update, repo_mirror_push, repo_topics_update, release_create
-- `delete_repo` — repo_delete
-
-Check current token: `curl -H "Authorization: token $GITEA_TOKEN" https://gitea.d-ma.be/api/v1/user`
-If scopes are missing, update token in Gitea settings before running tests.
-
-### Definition of done
-
-- `task check` passes (all tools, all batches)
-- Each new tool manually callable via `claude mcp call`
-- PR #1 (batch 1) merged before starting batch 2
-- Issue #19 (mirror flow e2e test) verified manually after batch 1 is deployed
+2. **Issue #19** — end-to-end mirror flow verification.
+   `repo_mirror_push` is implemented but the full flow (create repo → add push mirror → verify sync to GitHub) has not been tested manually. Do this before relying on it in production.

.gitea/workflows/cd.yml

@@ -4,8 +4,6 @@ on:
   push:
     branches: [main]
     tags: ["v*"]
-  pull_request:
-    branches: [main]
 
 env:
   IMAGE: gitea-mcp
@@ -43,7 +41,6 @@ jobs:
     name: Build & Import
     needs: check
     runs-on: self-hosted
-    if: github.event_name != 'pull_request'
     outputs:
       image-tag: ${{ steps.meta.outputs.sha-tag }}
     steps:

.githooks/pre-push

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+set -euo pipefail
+echo "→ Running task check before push..."
+task check
+echo "✓ pre-push check passed"

AGENTS.md

@@ -36,9 +36,18 @@ These rules apply to every task across every project, regardless of harness.
 4. **Goal-driven execution.** Define clear success criteria up front for every task.
    Loop — implement, verify, refine — until those criteria are met. Don't claim
    completion without evidence (tests pass, command output, observed behavior).
-5. **Branch-per-task for multi-agent repos.** When another agent may be active on
-   the same repo, create a branch (`agent/<description>`), commit there, and open a
-   PR. Do not merge without explicit instruction from Mathias.
+5. **Trunk-Based Development — commit directly to main.** Every commit is one
+   logical change (one tool, one fix, one test) with passing tests. Main is always
+   deployable. Never create long-lived feature branches.
+
+   **Exception — parallel agents on same repo:** If another agent is known to be
+   actively working on the same repo simultaneously, create a short-lived branch
+   (`agent/<description>`), finish the task, and merge to main within the same
+   session. Do not leave agent branches open between sessions.
+
+   **Exception — external contributor or client four-eyes requirement:** Use
+   PR flow only when a human reviewer outside the project is required. Document
+   the reason in PROJECT.md.
 
 ## Default stack
 
@@ -62,7 +71,10 @@ Exploratory: Rust, Zig — I'll tell you when I want these.
 - **Errors**: `fmt.Errorf("operation: %w", err)` — never naked, never log-and-return
 - **Naming**: stdlib conventions, no stuttering
 - **Architecture**: prefer stdlib over frameworks, constructor injection, env-var config parsed into typed structs
-- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), one concern per PR, PR describes *why* not *what*
+- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), commit directly to main,
+  one logical change per commit, CI is the quality gate
+- **Never**: long-lived feature branches, PRs for solo work, direct push without
+  passing `task check` locally first
 - **Security**: no secrets in code, govulncheck before adding deps, SOPS for encrypted config
 - **Dependencies**: prefer stdlib. testify, slog, templ, sqlc, google.golang.org/adk (agent projects only) are pre-approved; anything else needs justification in the commit message
 
@@ -104,18 +116,64 @@ See `~/dev/PROJECT_SUMMARY.md` for detailed descriptions of each project.
 - **koala-ai-stack** (`AGENTS/`) — local AI server infrastructure management
 - **klimatkollen** (`XT/`) — Swedish municipal climate data platform
 
-## Knowledge base
+## Knowledge base — actively use it
 
-When available, agents can query the shared knowledge base:
+A persistent brain (BM25 search + LLM-synthesised Q&A) survives across sessions,
+hosts, and harnesses. It holds 100+ hard-won entries: infra incident postmortems,
+Go pitfalls, framework gotchas, design principles, ADRs. **It is not optional
+reference material — query it actively, not just when explicitly told.**
 
-- **MCP**: `mcp://hyperguild.<TAILNET>.ts.net:3100/knowledge`
-- **HTTP**: `http://hyperguild.<TAILNET>.ts.net:3100/api/v1/search`
+### When to query (treat as a reflex)
 
-<!-- TODO: replace <TAILNET> placeholder with the real Tailscale tailnet
-     name once hyperguild is deployed. Until then, agents that try to
-     reach the knowledge service on a host where it isn't running will
-     get DNS NXDOMAIN, which is the desired fail-loudly behavior. -->
-- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`
+- **Before** starting a non-trivial task — search for prior art with the symptom
+  AND the system component ("how did we solve X in Y?"). 5 seconds beats 5 hours.
+- **When debugging** — search for the error string, the stack frame, the affected
+  service. Past you may have already paid this tax.
+- **Before adopting** a pattern, library, framework, or model name — check if it
+  was tried and rejected, or what the integration footguns are.
+- **When making architectural decisions** — search for the domain + "ADR" or
+  "decision" to find prior reasoning before re-deriving it.
+- **When a recommendation feels novel** — challenge yourself: "has this been
+  documented?" The brain often has it.
+
+### When to write
+
+After you discover something that **future-you would forget** and that **isn't
+recoverable from the code, git log, or PR description alone**:
+
+- Bugs whose root cause is non-obvious and generalisable beyond this project.
+- Framework / library / model-name quirks that bit you and would bite anyone.
+- Design principles validated under fire (e.g. "every `_get` needs a `_list`").
+- Postmortems for incidents: what broke, why, how diagnosed, what to do next time.
+
+DON'T write project status, sprint progress, PR summaries, or "what I did this
+session" — those rot fast and the originals are in git/gitea anyway. Brain
+entries that age well are about *why*, *how to avoid*, and *what to do when*.
+
+### How to access (per harness)
+
+| Harness | Query | Write |
+|---------|-------|-------|
+| **Claude Code, Claude Desktop** | `brain_query` (BM25), `brain_answer` (LLM-synth + sources) MCP tools | `brain_write` MCP tool |
+| **Crush, Pi, Antigravity, other MCP-capable** | same MCP server: `ingestion-brain` (via the `mcp__*_brain__*` namespace once authenticated) | same |
+| **Anything HTTP-only (curl, scripts)** | `POST https://brain-mcp.d-ma.be/query` with `{"query":"..."}` (auth via `BRAIN_MCP_TOKEN`) | `POST .../write` with `{"content":"...","filename":"..."}` |
+| **Browser / human inspection** | `https://gitea.d-ma.be/mathias/hyperguild` → `knowledge/` and `wiki/` markdown files |
+
+- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`.
+- **Routing**: brain_answer's LLM uses berget.ai as primary, iguana ollama as
+  fallback. Both are configurable in the `supervisor/ingestion-deployment.yaml`
+  on the koala k3s cluster; don't hardcode local-only model names into the
+  berget URL (see knowledge entry on namespace mismatches).
+
+### Quick reflex checks
+
+If you find yourself about to say any of these out loud, you owe yourself a brain query first:
+
+- "I think the issue might be..."
+- "Let me try X and see..."
+- "I'll just write a script to..."
+- "This is probably a new bug..."
+- "Has anyone done this before?" — *yes, probably, go check.*
 
 ## Client work rules
 
@@ -212,8 +270,11 @@ Key skills:
 
 ### Git
 - Conventional commits: `feat:`, `fix:`, `chore:`, `docs:`, `refactor:`
-- Branch naming: `feat/short-description`, `fix/short-description`
-- PRs: one concern per PR, description explains *why* not *what*
+- **Trunk-Based Development:** commit directly to main. One logical change per commit.
+- Run `task check` locally before every push. CI is the quality gate, not branch protection.
+- No feature branches, no PRs for solo/agent work.
+- Exception: if a parallel agent session is active on this repo, use a short-lived
+  `agent/<description>` branch and merge within the same session.
 
 ### Security
 - No secrets in code, ever — use env vars or SOPS-encrypted files
@@ -251,68 +312,29 @@ When acting as a coding agent on this project:
 3. If unsure about a convention, check `DECISIONS.md` or ask
 4. Never modify files outside the project root without explicit permission
 5. When adding a dependency, explain why in the commit message
-6. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
+6. Commit directly to main. Run `task check` before every push. Never create
+   feature branches unless a parallel agent is simultaneously active on this repo.
+7. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
 
-## Current sprint — gitea-mcp v0.2 (2026-05-14)
+## Current state — v0.2.5 (2026-05-17)
 
-### Context
-This sprint implements new MCP tools needed for `hyperguild new-project` —
-the automated project creation flow triggered from claude.ai. See brain knowledge
-nodes `adr-new-project-gitea-first-github-mirror` and `roadmap-github-ingestion-pipeline`
-for full background.
+All v0.2 work is complete and deployed. No active sprint.
 
-### Issues to implement (priority order)
+### What shipped
 
-**Batch 1 — blockers (do first, one PR: `feat/repo-crud`)**
+| Tag | PR | Tools / fixes |
+|-----|----|---------------|
+| v0.2.2 | #21 | repo_create, repo_update, repo_mirror_push |
+| v0.2.3 | #22 | repo_tree, repo_topics_update, file_read dir fix |
+| v0.2.4 | #23 | issue_get, release_create, repo_delete |
+| v0.2.5 | #26 | repo_update archived+template, create_project_from_template template_name, pr_files_diff loop fix |
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #13 | `repo_create` | POST /api/v1/user/repos or /api/v1/orgs/{org}/repos |
-| #16 | `repo_mirror_push` (add/list/delete) | POST/GET/DELETE /api/v1/repos/{owner}/{repo}/push_mirrors |
-| #12 | `repo_update` | PATCH /api/v1/repos/{owner}/{repo} |
+Current main: `e31fd3f`. CI green. Deployed via Flux.
 
-**Batch 2 — quality of life (second PR: `feat/repo-ux`)**
+### Next up
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #15 | `file_read` dir-path fix | existing endpoint, detect array vs object response |
-| #14 | `repo_tree` | GET /api/v1/repos/{owner}/{repo}/git/trees/{sha}?recursive=true |
-| #18 | `repo_topics_update` | PUT /api/v1/repos/{owner}/{repo}/topics |
+1. **`hyperguild new-project` v1** — primary next target.
+   See brain node `adr-new-project-gitea-first-github-mirror` for full flow spec.
 
-**Batch 3 — can wait**
-
-| Issue | Tool | Note |
-|-------|------|------|
-| #11 | `repo_delete` | HIGH risk — needs `confirm` param == repo name |
-| #17 | `release_create` | POST /api/v1/repos/{owner}/{repo}/releases |
-
-### How to add a tool (pattern)
-
-Every tool = 4 files following `internal/tools/repo_get.go` exactly:
-
-1. `internal/gitea/<domain>.go` — API client method (use PostJSON/PatchJSON/DeleteJSON)
-2. `internal/tools/repo_<name>.go` — tool handler with Descriptor() + Call()
-3. `internal/tools/repo_<name>_test.go` — table-driven tests with httptest.NewServer
-4. Registration in main — find where `NewRepoGet` is registered, add new tool same place
-
-Key rules:
-- Always call `t.a.Check(args.Owner)` before any API call (allowlist guard)
-- Use `textOK(result)` for success output
-- For `repo_mirror_push`: NEVER log or return `remote_password` in any output
-- For `repo_update` with `private: false` and `repo_delete`: require `confirm` param == repo name
-
-### Token permissions needed
-
-New tools require these additional Gitea token scopes:
-- `write:repository` — repo_create, repo_update, repo_mirror_push, repo_topics_update, release_create
-- `delete_repo` — repo_delete
-
-Check current token: `curl -H "Authorization: token $GITEA_TOKEN" https://gitea.d-ma.be/api/v1/user`
-If scopes are missing, update token in Gitea settings before running tests.
-
-### Definition of done
-
-- `task check` passes (all tools, all batches)
-- Each new tool manually callable via `claude mcp call`
-- PR #1 (batch 1) merged before starting batch 2
-- Issue #19 (mirror flow e2e test) verified manually after batch 1 is deployed
+2. **Issue #19** — end-to-end mirror flow verification.
+   `repo_mirror_push` is implemented but the full flow (create repo → add push mirror → verify sync to GitHub) has not been tested manually. Do this before relying on it in production.

CLAUDE.md

@@ -37,8 +37,11 @@
 
 ### Git
 - Conventional commits: `feat:`, `fix:`, `chore:`, `docs:`, `refactor:`
-- Branch naming: `feat/short-description`, `fix/short-description`
-- PRs: one concern per PR, description explains *why* not *what*
+- **Trunk-Based Development:** commit directly to main. One logical change per commit.
+- Run `task check` locally before every push. CI is the quality gate, not branch protection.
+- No feature branches, no PRs for solo/agent work.
+- Exception: if a parallel agent session is active on this repo, use a short-lived
+  `agent/<description>` branch and merge within the same session.
 
 ### Security
 - No secrets in code, ever — use env vars or SOPS-encrypted files
@@ -76,68 +79,29 @@ When acting as a coding agent on this project:
 3. If unsure about a convention, check `DECISIONS.md` or ask
 4. Never modify files outside the project root without explicit permission
 5. When adding a dependency, explain why in the commit message
-6. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
+6. Commit directly to main. Run `task check` before every push. Never create
+   feature branches unless a parallel agent is simultaneously active on this repo.
+7. For client projects: never send code or context to cloud APIs — use local models via LiteLLM
 
-## Current sprint — gitea-mcp v0.2 (2026-05-14)
+## Current state — v0.2.5 (2026-05-17)
 
-### Context
-This sprint implements new MCP tools needed for `hyperguild new-project` —
-the automated project creation flow triggered from claude.ai. See brain knowledge
-nodes `adr-new-project-gitea-first-github-mirror` and `roadmap-github-ingestion-pipeline`
-for full background.
+All v0.2 work is complete and deployed. No active sprint.
 
-### Issues to implement (priority order)
+### What shipped
 
-**Batch 1 — blockers (do first, one PR: `feat/repo-crud`)**
+| Tag | PR | Tools / fixes |
+|-----|----|---------------|
+| v0.2.2 | #21 | repo_create, repo_update, repo_mirror_push |
+| v0.2.3 | #22 | repo_tree, repo_topics_update, file_read dir fix |
+| v0.2.4 | #23 | issue_get, release_create, repo_delete |
+| v0.2.5 | #26 | repo_update archived+template, create_project_from_template template_name, pr_files_diff loop fix |
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #13 | `repo_create` | POST /api/v1/user/repos or /api/v1/orgs/{org}/repos |
-| #16 | `repo_mirror_push` (add/list/delete) | POST/GET/DELETE /api/v1/repos/{owner}/{repo}/push_mirrors |
-| #12 | `repo_update` | PATCH /api/v1/repos/{owner}/{repo} |
+Current main: `e31fd3f`. CI green. Deployed via Flux.
 
-**Batch 2 — quality of life (second PR: `feat/repo-ux`)**
+### Next up
 
-| Issue | Tool | Gitea API |
-|-------|------|-----------|
-| #15 | `file_read` dir-path fix | existing endpoint, detect array vs object response |
-| #14 | `repo_tree` | GET /api/v1/repos/{owner}/{repo}/git/trees/{sha}?recursive=true |
-| #18 | `repo_topics_update` | PUT /api/v1/repos/{owner}/{repo}/topics |
+1. **`hyperguild new-project` v1** — primary next target.
+   See brain node `adr-new-project-gitea-first-github-mirror` for full flow spec.
 
-**Batch 3 — can wait**
-
-| Issue | Tool | Note |
-|-------|------|------|
-| #11 | `repo_delete` | HIGH risk — needs `confirm` param == repo name |
-| #17 | `release_create` | POST /api/v1/repos/{owner}/{repo}/releases |
-
-### How to add a tool (pattern)
-
-Every tool = 4 files following `internal/tools/repo_get.go` exactly:
-
-1. `internal/gitea/<domain>.go` — API client method (use PostJSON/PatchJSON/DeleteJSON)
-2. `internal/tools/repo_<name>.go` — tool handler with Descriptor() + Call()
-3. `internal/tools/repo_<name>_test.go` — table-driven tests with httptest.NewServer
-4. Registration in main — find where `NewRepoGet` is registered, add new tool same place
-
-Key rules:
-- Always call `t.a.Check(args.Owner)` before any API call (allowlist guard)
-- Use `textOK(result)` for success output
-- For `repo_mirror_push`: NEVER log or return `remote_password` in any output
-- For `repo_update` with `private: false` and `repo_delete`: require `confirm` param == repo name
-
-### Token permissions needed
-
-New tools require these additional Gitea token scopes:
-- `write:repository` — repo_create, repo_update, repo_mirror_push, repo_topics_update, release_create
-- `delete_repo` — repo_delete
-
-Check current token: `curl -H "Authorization: token $GITEA_TOKEN" https://gitea.d-ma.be/api/v1/user`
-If scopes are missing, update token in Gitea settings before running tests.
-
-### Definition of done
-
-- `task check` passes (all tools, all batches)
-- Each new tool manually callable via `claude mcp call`
-- PR #1 (batch 1) merged before starting batch 2
-- Issue #19 (mirror flow e2e test) verified manually after batch 1 is deployed
+2. **Issue #19** — end-to-end mirror flow verification.
+   `repo_mirror_push` is implemented but the full flow (create repo → add push mirror → verify sync to GitHub) has not been tested manually. Do this before relying on it in production.

README.md

@@ -2,3 +2,14 @@
 
 Streamable HTTP MCP service exposing Gitea repo operations to Claude apps.
 See `~/dev/AI/infra/docs/superpowers/specs/2026-05-04-gitea-mcp-gitops-workflow-design.md`.
+
+## Quickstart
+
+```bash
+task setup:hooks   # installs .githooks/pre-push — runs task check before every push
+task check         # context sync + lint + test + vet
+task build         # produces bin/gitea-mcp
+```
+
+This repo uses Trunk-Based Development. Commit directly to `main`. The pre-push
+hook enforces the quality gate locally; CI re-runs `task check` on every push.

Taskfile.yml

@@ -47,6 +47,13 @@ tasks:
     cmds:
       - bash scripts/context-sync.sh
 
+  setup:hooks:
+    desc: Install git hooks (.githooks/pre-push)
+    cmds:
+      - git config core.hooksPath .githooks
+      - chmod +x .githooks/pre-push
+      - echo "✓ git hooks installed (pre-push runs task check)"
+
   context:sync:claude:
     cmds: [bash scripts/context-sync.sh claude]
   context:sync:agents:

cmd/gitea-mcp/main.go

@@ -66,6 +66,8 @@ func main() {
 	reg.Register(tools.NewRepoTree(giteaClient, ownerAllow))
 	reg.Register(tools.NewRepoTopicsUpdate(giteaClient, ownerAllow))
 	reg.Register(tools.NewIssueGet(giteaClient, ownerAllow))
+	reg.Register(tools.NewIssueClose(giteaClient, ownerAllow))
+	reg.Register(tools.NewIssueReopen(giteaClient, ownerAllow))
 	reg.Register(tools.NewReleaseCreate(giteaClient, ownerAllow))
 	reg.Register(tools.NewRepoDelete(giteaClient, ownerAllow))
 

internal/gitea/issues.go

@@ -72,6 +72,28 @@ func (c *Client) CreateIssue(ctx context.Context, owner, repo string, args Creat
 	return &iss, nil
 }
 
+// SetIssueState flips an issue between "open" and "closed" via PATCH.
+// Gitea uses the same endpoint for both transitions.
+func (c *Client) SetIssueState(ctx context.Context, owner, repo string, number int, state string) (*Issue, error) {
+	p := fmt.Sprintf("/api/v1/repos/%s/%s/issues/%d", owner, repo, number)
+	payload, err := json.Marshal(map[string]string{"state": state})
+	if err != nil {
+		return nil, err
+	}
+	body, status, err := c.PatchJSON(ctx, p, payload)
+	if err != nil {
+		return nil, err
+	}
+	if err := MapStatus(status, body); err != nil {
+		return nil, err
+	}
+	var iss Issue
+	if err := json.Unmarshal(body, &iss); err != nil {
+		return nil, err
+	}
+	return &iss, nil
+}
+
 type IssueComment struct {
 	ID      int64  `json:"id"`
 	Body    string `json:"body"`

internal/gitea/repos.go

@@ -216,6 +216,8 @@ type UpdateRepoArgs struct {
 	Private       *bool   `json:"private,omitempty"`
 	Website       *string `json:"website,omitempty"`
 	DefaultBranch *string `json:"default_branch,omitempty"`
+	Archived      *bool   `json:"archived,omitempty"`
+	Template      *bool   `json:"template,omitempty"`
 }
 
 func (c *Client) UpdateRepo(ctx context.Context, owner, name string, args UpdateRepoArgs) (*Repo, error) {
@@ -253,3 +255,4 @@ func (c *Client) GetRepo(ctx context.Context, owner, name string) (*Repo, error)
 	}
 	return &r, nil
 }
+

internal/tools/create_project_from_template.go

@@ -45,14 +45,15 @@ func NewCreateProjectFromTemplate(c *gitea.Client, a *allowlist.Allowlist, tmplO
 func (t *CreateProjectFromTemplate) Descriptor() registry.ToolDescriptor {
 	return registry.ToolDescriptor{
 		Name:        "create_project_from_template",
-		Description: "Create a new project repo from the template, applying placeholder substitutions to known files.",
+		Description: "Create a new project repo from a template, applying placeholder substitutions to known files. Defaults to the server-configured template; pass template_name to override (e.g. template-go-agent).",
 		InputSchema: json.RawMessage(`{
 			"type":"object",
 			"properties":{
 				"owner":{"type":"string"},
 				"name":{"type":"string","pattern":"^[a-z][a-z0-9-]{1,38}[a-z0-9]$"},
 				"description":{"type":"string"},
-				"private":{"type":"boolean"}
+				"private":{"type":"boolean"},
+				"template_name":{"type":"string","description":"Template repo name to generate from. Defaults to the server-configured template."}
 			},
 			"required":["owner","name"]
 		}`),
@@ -64,6 +65,7 @@ type createProjectArgs struct {
 	Name         string `json:"name"`
 	Description  string `json:"description"`
 	Private      bool   `json:"private"`
+	TemplateName string `json:"template_name"`
 }
 
 type createProjectResult struct {
@@ -91,13 +93,20 @@ func (t *CreateProjectFromTemplate) Call(ctx context.Context, raw json.RawMessag
 		return nil, fmt.Errorf("name %q does not match pattern %s: %w", args.Name, nameRe.String(), gitea.ErrValidation)
 	}
 
+	// Resolve template: per-call override takes precedence over the
+	// server-configured default. Owner stays server-configured.
+	tmplName := args.TemplateName
+	if tmplName == "" {
+		tmplName = t.templateName
+	}
+
 	// Verify template exists and is marked as a template repo.
-	tmpl, err := t.c.GetRepo(ctx, t.templateOwner, t.templateName)
+	tmpl, err := t.c.GetRepo(ctx, t.templateOwner, tmplName)
 	if err != nil {
 		return nil, fmt.Errorf("template lookup: %w", err)
 	}
 	if !tmpl.Template {
-		return nil, fmt.Errorf("repo %s/%s is not marked as template: %w", t.templateOwner, t.templateName, gitea.ErrValidation)
+		return nil, fmt.Errorf("repo %s/%s is not marked as template: %w", t.templateOwner, tmplName, gitea.ErrValidation)
 	}
 
 	// Verify destination doesn't already exist.
@@ -108,7 +117,7 @@ func (t *CreateProjectFromTemplate) Call(ctx context.Context, raw json.RawMessag
 	}
 
 	// Generate repo from template.
-	newRepo, err := t.c.GenerateFromTemplate(ctx, t.templateOwner, t.templateName, gitea.GenerateFromTemplateArgs{
+	newRepo, err := t.c.GenerateFromTemplate(ctx, t.templateOwner, tmplName, gitea.GenerateFromTemplateArgs{
 		Owner:       args.Owner,
 		Name:        args.Name,
 		Description: args.Description,

internal/tools/create_project_from_template_test.go

@@ -122,6 +122,62 @@ func TestCreateProjectHappyPath(t *testing.T) {
 	assert.Empty(t, out.PartialFailure)
 }
 
+// TestCreateProjectTemplateNameOverride (issue #24): per-call template_name overrides the
+// server-configured default, so the same binary can generate from template-go-web or
+// template-go-agent without restart.
+func TestCreateProjectTemplateNameOverride(t *testing.T) {
+	var templateLookups, generateCalls []string
+
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "application/json")
+		switch {
+		case r.Method == http.MethodGet && r.URL.Path == "/api/v1/repos/mathias/template-go-agent":
+			templateLookups = append(templateLookups, "template-go-agent")
+			_, _ = w.Write([]byte(newTemplateRepoJSON("template-go-agent", true)))
+
+		case r.Method == http.MethodGet && r.URL.Path == "/api/v1/repos/mathias/template-go-web":
+			templateLookups = append(templateLookups, "template-go-web")
+			_, _ = w.Write([]byte(newTemplateRepoJSON("template-go-web", true)))
+
+		case r.Method == http.MethodGet && r.URL.Path == "/api/v1/repos/mathias/new-agent":
+			w.WriteHeader(http.StatusNotFound)
+			_, _ = w.Write([]byte(`{"message":"not found"}`))
+
+		case r.Method == http.MethodPost && strings.HasSuffix(r.URL.Path, "/generate"):
+			generateCalls = append(generateCalls, r.URL.Path)
+			w.WriteHeader(http.StatusCreated)
+			_, _ = w.Write([]byte(newGeneratedRepoJSON("new-agent")))
+
+		case r.Method == http.MethodGet && strings.HasPrefix(r.URL.Path, "/api/v1/repos/mathias/new-agent/contents/"):
+			filePath := strings.TrimPrefix(r.URL.Path, "/api/v1/repos/mathias/new-agent/contents/")
+			_, _ = w.Write([]byte(fileContentsJSON(filePath)))
+
+		case r.Method == http.MethodPut && strings.HasPrefix(r.URL.Path, "/api/v1/repos/mathias/new-agent/contents/"):
+			filePath := strings.TrimPrefix(r.URL.Path, "/api/v1/repos/mathias/new-agent/contents/")
+			w.WriteHeader(http.StatusOK)
+			_, _ = w.Write([]byte(fileWriteResultJSON(filePath)))
+
+		default:
+			t.Errorf("unexpected request: %s %s", r.Method, r.URL.Path)
+			w.WriteHeader(http.StatusNotFound)
+		}
+	}))
+	defer srv.Close()
+
+	// Server is configured with template-go-web as the default; call overrides to template-go-agent.
+	tool := newCreateProjectTool(srv.URL)
+	_, err := tool.Call(context.Background(), json.RawMessage(
+		`{"owner":"mathias","name":"new-agent","template_name":"template-go-agent"}`,
+	))
+	require.NoError(t, err)
+
+	assert.Equal(t, []string{"template-go-agent"}, templateLookups,
+		"override must direct the template lookup, not the server default")
+	require.Len(t, generateCalls, 1)
+	assert.Equal(t, "/api/v1/repos/mathias/template-go-agent/generate", generateCalls[0],
+		"override must direct the /generate call too")
+}
+
 // TestCreateProjectNameRegexFailure: invalid name returns ErrValidation without hitting network.
 func TestCreateProjectNameRegexFailure(t *testing.T) {
 	tool := tools.NewCreateProjectFromTemplate(

internal/tools/issue_close.go

@@ -0,0 +1,56 @@
+package tools
+
+import (
+	"context"
+	"encoding/json"
+
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/allowlist"
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/gitea"
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/registry"
+)
+
+type IssueClose struct {
+	c *gitea.Client
+	a *allowlist.Allowlist
+}
+
+func NewIssueClose(c *gitea.Client, a *allowlist.Allowlist) *IssueClose {
+	return &IssueClose{c: c, a: a}
+}
+
+func (t *IssueClose) Descriptor() registry.ToolDescriptor {
+	return registry.ToolDescriptor{
+		Name:        "issue_close",
+		Description: "Close an open issue. Reversible via issue_reopen.",
+		InputSchema: json.RawMessage(`{
+			"type":"object",
+			"properties":{
+				"owner":{"type":"string"},
+				"name":{"type":"string"},
+				"number":{"type":"integer","minimum":1}
+			},
+			"required":["owner","name","number"]
+		}`),
+	}
+}
+
+type issueCloseArgs struct {
+	Owner  string `json:"owner"`
+	Name   string `json:"name"`
+	Number int    `json:"number"`
+}
+
+func (t *IssueClose) Call(ctx context.Context, raw json.RawMessage) (json.RawMessage, error) {
+	var args issueCloseArgs
+	if err := parseArgs(raw, &args); err != nil {
+		return nil, err
+	}
+	if err := t.a.Check(args.Owner); err != nil {
+		return nil, err
+	}
+	iss, err := t.c.SetIssueState(ctx, args.Owner, args.Name, args.Number, "closed")
+	if err != nil {
+		return nil, err
+	}
+	return textOK(iss)
+}

internal/tools/issue_close_test.go

@@ -0,0 +1,52 @@
+package tools_test
+
+import (
+	"context"
+	"encoding/json"
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/allowlist"
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/gitea"
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/tools"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestIssueCloseTool(t *testing.T) {
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		assert.Equal(t, http.MethodPatch, r.Method)
+		assert.Equal(t, "/api/v1/repos/mathias/infra/issues/26", r.URL.Path)
+		b, _ := io.ReadAll(r.Body)
+		assert.JSONEq(t, `{"state":"closed"}`, string(b))
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{"number":26,"title":"feat: ntfy via NPM","state":"closed","html_url":"http://gitea.example.com/mathias/infra/issues/26"}`))
+	}))
+	defer srv.Close()
+
+	tool := tools.NewIssueClose(gitea.NewClient(srv.URL, "tok"), allowlist.New([]string{"mathias"}))
+	out, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"mathias","name":"infra","number":26}`))
+	require.NoError(t, err)
+	assert.Contains(t, string(out), `"number":26`)
+	assert.Contains(t, string(out), `"state":"closed"`)
+}
+
+func TestIssueCloseTool_NotFound(t *testing.T) {
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.WriteHeader(http.StatusNotFound)
+		_, _ = w.Write([]byte(`{"message":"issue not found"}`))
+	}))
+	defer srv.Close()
+
+	tool := tools.NewIssueClose(gitea.NewClient(srv.URL, "tok"), allowlist.New([]string{"mathias"}))
+	_, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"mathias","name":"infra","number":999}`))
+	require.Error(t, err)
+}
+
+func TestIssueCloseAllowlistRejects(t *testing.T) {
+	tool := tools.NewIssueClose(gitea.NewClient("http://unused", ""), allowlist.New([]string{"mathias"}))
+	_, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"evil","name":"x","number":1}`))
+	require.Error(t, err)
+}

internal/tools/issue_reopen.go

@@ -0,0 +1,56 @@
+package tools
+
+import (
+	"context"
+	"encoding/json"
+
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/allowlist"
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/gitea"
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/registry"
+)
+
+type IssueReopen struct {
+	c *gitea.Client
+	a *allowlist.Allowlist
+}
+
+func NewIssueReopen(c *gitea.Client, a *allowlist.Allowlist) *IssueReopen {
+	return &IssueReopen{c: c, a: a}
+}
+
+func (t *IssueReopen) Descriptor() registry.ToolDescriptor {
+	return registry.ToolDescriptor{
+		Name:        "issue_reopen",
+		Description: "Reopen a closed issue. Reversible via issue_close.",
+		InputSchema: json.RawMessage(`{
+			"type":"object",
+			"properties":{
+				"owner":{"type":"string"},
+				"name":{"type":"string"},
+				"number":{"type":"integer","minimum":1}
+			},
+			"required":["owner","name","number"]
+		}`),
+	}
+}
+
+type issueReopenArgs struct {
+	Owner  string `json:"owner"`
+	Name   string `json:"name"`
+	Number int    `json:"number"`
+}
+
+func (t *IssueReopen) Call(ctx context.Context, raw json.RawMessage) (json.RawMessage, error) {
+	var args issueReopenArgs
+	if err := parseArgs(raw, &args); err != nil {
+		return nil, err
+	}
+	if err := t.a.Check(args.Owner); err != nil {
+		return nil, err
+	}
+	iss, err := t.c.SetIssueState(ctx, args.Owner, args.Name, args.Number, "open")
+	if err != nil {
+		return nil, err
+	}
+	return textOK(iss)
+}

internal/tools/issue_reopen_test.go

@@ -0,0 +1,40 @@
+package tools_test
+
+import (
+	"context"
+	"encoding/json"
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/allowlist"
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/gitea"
+	"gitea.d-ma.be/mathias/gitea-mcp/internal/tools"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestIssueReopenTool(t *testing.T) {
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		assert.Equal(t, http.MethodPatch, r.Method)
+		assert.Equal(t, "/api/v1/repos/mathias/infra/issues/26", r.URL.Path)
+		b, _ := io.ReadAll(r.Body)
+		assert.JSONEq(t, `{"state":"open"}`, string(b))
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{"number":26,"title":"feat: ntfy via NPM","state":"open","html_url":"http://gitea.example.com/mathias/infra/issues/26"}`))
+	}))
+	defer srv.Close()
+
+	tool := tools.NewIssueReopen(gitea.NewClient(srv.URL, "tok"), allowlist.New([]string{"mathias"}))
+	out, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"mathias","name":"infra","number":26}`))
+	require.NoError(t, err)
+	assert.Contains(t, string(out), `"number":26`)
+	assert.Contains(t, string(out), `"state":"open"`)
+}
+
+func TestIssueReopenAllowlistRejects(t *testing.T) {
+	tool := tools.NewIssueReopen(gitea.NewClient("http://unused", ""), allowlist.New([]string{"mathias"}))
+	_, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"evil","name":"x","number":1}`))
+	require.Error(t, err)
+}

internal/tools/pr_files_diff.go

@@ -143,7 +143,13 @@ func splitUnifiedDiff(d []byte) map[string][]byte {
 
 	flush := func() {
 		if currentFile != "" {
-			m[currentFile] = current.Bytes()
+			// Copy: bytes.Buffer.Bytes() returns the internal slice,
+			// which Reset() then reuses. Without the copy, every map
+			// entry ends up aliased to the last file's data.
+			b := current.Bytes()
+			cp := make([]byte, len(b))
+			copy(cp, b)
+			m[currentFile] = cp
 			current.Reset()
 		}
 	}

internal/tools/pr_files_diff_test.go

@@ -97,6 +97,47 @@ func TestPRFilesDiffSmall(t *testing.T) {
 	assert.ElementsMatch(t, fileNames, paths)
 }
 
+// Regression for issue #25: every file's diff entry must contain its OWN diff,
+// not a shared buffer pointing at the last file. Prior bug: splitUnifiedDiff
+// flushed bytes.Buffer.Bytes() into the map without copying, so every entry
+// aliased the buffer's backing array and showed the last file's content.
+func TestPRFilesDiffPerFileIsolation(t *testing.T) {
+	fileNames := []string{"alpha.go", "beta.go", "gamma.go", "delta.go"}
+	rawDiff := buildDiff(fileNames, 5)
+	filesJSON := buildFilesJSON(fileNames, 5)
+
+	srv := newPRFilesDiffServer(t, filesJSON, rawDiff)
+	defer srv.Close()
+
+	tool := tools.NewPRFilesDiff(gitea.NewClient(srv.URL, "tok"), allowlist.New([]string{"o"}))
+	result, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"o","name":"r","number":1}`))
+	require.NoError(t, err)
+
+	var out struct {
+		Files []struct {
+			Path string `json:"path"`
+			Diff string `json:"diff"`
+		} `json:"files"`
+	}
+	require.NoError(t, json.Unmarshal(result, &out))
+	require.Len(t, out.Files, len(fileNames))
+
+	for _, f := range out.Files {
+		expected := fmt.Sprintf("diff --git a/%s b/%s", f.Path, f.Path)
+		assert.Contains(t, f.Diff, expected,
+			"file %s diff must contain its own header, got: %.80q", f.Path, f.Diff)
+		// No other file's header should leak in.
+		for _, other := range fileNames {
+			if other == f.Path {
+				continue
+			}
+			otherHeader := fmt.Sprintf("diff --git a/%s b/%s", other, other)
+			assert.NotContains(t, f.Diff, otherHeader,
+				"file %s diff must NOT contain %s's header", f.Path, other)
+		}
+	}
+}
+
 func TestPRFilesDiffPerFileTruncated(t *testing.T) {
 	// One file with a 30KB diff (each "+abcdefghij\n" = 12 bytes; 30KB / 12 ≈ 2560 lines).
 	fileNames := []string{"bigfile.go"}

internal/tools/repo_update.go

@@ -22,16 +22,20 @@ func NewRepoUpdate(c *gitea.Client, a *allowlist.Allowlist) *RepoUpdate {
 func (t *RepoUpdate) Descriptor() registry.ToolDescriptor {
 	return registry.ToolDescriptor{
 		Name: "repo_update",
-		Description: "Update repository metadata (description, visibility, default branch, website).",
+		Description: "Update repository metadata (description, visibility, default branch, website, archived, template). " +
+			"Only fields explicitly set in the call are patched. " +
+			"WARNING: private=false exposes the repo publicly — verify intent before calling.",
 		InputSchema: json.RawMessage(`{
 			"type":"object",
 			"properties":{
 				"owner":{"type":"string"},
 				"name":{"type":"string"},
 				"description":{"type":"string"},
-				"private":{"type":"boolean"},
-				"website":{"type":"string"},
-				"default_branch":{"type":"string"},
+				"private":{"type":"boolean","description":"Toggle visibility. false makes the repo public."},
+				"website":{"type":"string","description":"Homepage URL"},
+				"default_branch":{"type":"string","description":"Rename the default branch"},
+				"archived":{"type":"boolean","description":"Mark repo as archived (read-only)."},
+				"template":{"type":"boolean","description":"Toggle template-repo flag"},
 				"confirm":{"type":"string","description":"Required when setting private=false. Must equal the repo name."}
 			},
 			"required":["owner","name"]
@@ -42,10 +46,12 @@ func (t *RepoUpdate) Descriptor() registry.ToolDescriptor {
 type repoUpdateArgs struct {
 	Owner         string  `json:"owner"`
 	Name          string  `json:"name"`
-	Description   *string `json:"description"`
-	Private       *bool   `json:"private"`
-	Website       *string `json:"website"`
-	DefaultBranch *string `json:"default_branch"`
+	Description   *string `json:"description,omitempty"`
+	Private       *bool   `json:"private,omitempty"`
+	Website       *string `json:"website,omitempty"`
+	DefaultBranch *string `json:"default_branch,omitempty"`
+	Archived      *bool   `json:"archived,omitempty"`
+	Template      *bool   `json:"template,omitempty"`
 	Confirm       string  `json:"confirm"`
 }
 
@@ -57,17 +63,26 @@ func (t *RepoUpdate) Call(ctx context.Context, raw json.RawMessage) (json.RawMes
 	if err := t.a.Check(args.Owner); err != nil {
 		return nil, err
 	}
+
 	// Making a repo public is a significant action — require explicit confirmation.
 	if args.Private != nil && !*args.Private {
 		if args.Confirm != args.Name {
 			return nil, fmt.Errorf("setting private=false makes the repo public: set confirm=%q to proceed", args.Name)
 		}
 	}
+
+	if args.Description == nil && args.Private == nil && args.Website == nil &&
+		args.DefaultBranch == nil && args.Archived == nil && args.Template == nil {
+		return nil, fmt.Errorf("at least one updatable field must be set: %w", gitea.ErrValidation)
+	}
+
 	r, err := t.c.UpdateRepo(ctx, args.Owner, args.Name, gitea.UpdateRepoArgs{
 		Description:   args.Description,
 		Private:       args.Private,
 		Website:       args.Website,
 		DefaultBranch: args.DefaultBranch,
+		Archived:      args.Archived,
+		Template:      args.Template,
 	})
 	if err != nil {
 		return nil, err

internal/tools/repo_update_test.go

@@ -3,6 +3,7 @@ package tools_test
 import (
 	"context"
 	"encoding/json"
+	"io"
 	"net/http"
 	"net/http/httptest"
 	"testing"
@@ -14,43 +15,139 @@ import (
 	"github.com/stretchr/testify/require"
 )
 
-func TestRepoUpdateTool(t *testing.T) {
+func newRepoUpdateTool(srvURL string) *tools.RepoUpdate {
+	return tools.NewRepoUpdate(gitea.NewClient(srvURL, "tok"), allowlist.New([]string{"mathias"}))
+}
+
+// TestRepoUpdateArchive: happy path — set archived=true.
+func TestRepoUpdateArchive(t *testing.T) {
+	var patchedBody []byte
 	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
-		assert.Equal(t, http.MethodPatch, r.Method)
-		assert.Equal(t, "/api/v1/repos/mathias/infra", r.URL.Path)
+		require.Equal(t, http.MethodPatch, r.Method)
+		require.Equal(t, "/api/v1/repos/mathias/old-svc", r.URL.Path)
+		patchedBody, _ = io.ReadAll(r.Body)
 		w.Header().Set("Content-Type", "application/json")
-		_, _ = w.Write([]byte(`{"name":"infra","full_name":"mathias/infra","default_branch":"main","description":"updated","private":true,"clone_url":"https://gitea.example.com/mathias/infra.git","html_url":"https://gitea.example.com/mathias/infra"}`))
+		_, _ = w.Write([]byte(`{"name":"old-svc","full_name":"mathias/old-svc","default_branch":"main","template":false,"private":false}`))
 	}))
 	defer srv.Close()
 
-	tool := tools.NewRepoUpdate(gitea.NewClient(srv.URL, "tok"), allowlist.New([]string{"mathias"}))
-	out, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"mathias","name":"infra","description":"updated"}`))
+	tool := newRepoUpdateTool(srv.URL)
+	result, err := tool.Call(context.Background(), json.RawMessage(
+		`{"owner":"mathias","name":"old-svc","archived":true}`,
+	))
 	require.NoError(t, err)
-	assert.Contains(t, string(out), `"description":"updated"`)
+
+	// Wire payload only contains the field that was actually set.
+	var sent map[string]any
+	require.NoError(t, json.Unmarshal(patchedBody, &sent))
+	assert.Equal(t, true, sent["archived"])
+	assert.NotContains(t, sent, "description")
+	assert.NotContains(t, sent, "private")
+	assert.NotContains(t, sent, "website")
+	assert.NotContains(t, sent, "template")
+
+	var repo gitea.Repo
+	require.NoError(t, json.Unmarshal(result, &repo))
+	assert.Equal(t, "mathias/old-svc", repo.FullName)
 }
 
-func TestRepoUpdateTool_MakePublicRequiresConfirm(t *testing.T) {
-	tool := tools.NewRepoUpdate(gitea.NewClient("http://unused", ""), allowlist.New([]string{"mathias"}))
-	_, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"mathias","name":"infra","private":false}`))
+// TestRepoUpdateMultipleFields: set description + template flag in one call.
+func TestRepoUpdateMultipleFields(t *testing.T) {
+	var patchedBody []byte
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		patchedBody, _ = io.ReadAll(r.Body)
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{"name":"template-go-agent","full_name":"mathias/template-go-agent","description":"Go agent template","template":true}`))
+	}))
+	defer srv.Close()
+
+	tool := newRepoUpdateTool(srv.URL)
+	_, err := tool.Call(context.Background(), json.RawMessage(
+		`{"owner":"mathias","name":"template-go-agent","description":"Go agent template","template":true}`,
+	))
+	require.NoError(t, err)
+
+	var sent map[string]any
+	require.NoError(t, json.Unmarshal(patchedBody, &sent))
+	assert.Equal(t, "Go agent template", sent["description"])
+	assert.Equal(t, true, sent["template"])
+	assert.NotContains(t, sent, "archived")
+	assert.NotContains(t, sent, "private")
+}
+
+// TestRepoUpdateNoFieldsRejected: zero updatable fields → validation error before network.
+func TestRepoUpdateNoFieldsRejected(t *testing.T) {
+	tool := tools.NewRepoUpdate(
+		gitea.NewClient("http://unused", ""),
+		allowlist.New([]string{"mathias"}),
+	)
+	_, err := tool.Call(context.Background(), json.RawMessage(
+		`{"owner":"mathias","name":"some-repo"}`,
+	))
+	require.Error(t, err)
+	assert.ErrorIs(t, err, gitea.ErrValidation)
+}
+
+// TestRepoUpdateMakePublic: private=false requires confirm=<repo name> as a safety
+// gate (kept from main #21 during the v02-patch merge). With confirm matching, the
+// patch goes through.
+func TestRepoUpdateMakePublic(t *testing.T) {
+	var patchedBody []byte
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		patchedBody, _ = io.ReadAll(r.Body)
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{"name":"open-repo","full_name":"mathias/open-repo","private":false}`))
+	}))
+	defer srv.Close()
+
+	tool := newRepoUpdateTool(srv.URL)
+	_, err := tool.Call(context.Background(), json.RawMessage(
+		`{"owner":"mathias","name":"open-repo","private":false,"confirm":"open-repo"}`,
+	))
+	require.NoError(t, err)
+
+	var sent map[string]any
+	require.NoError(t, json.Unmarshal(patchedBody, &sent))
+	assert.Equal(t, false, sent["private"])
+}
+
+// TestRepoUpdateMakePublicWithoutConfirm: confirm gate blocks private=false without confirmation.
+func TestRepoUpdateMakePublicWithoutConfirm(t *testing.T) {
+	tool := tools.NewRepoUpdate(
+		gitea.NewClient("http://unused", ""),
+		allowlist.New([]string{"mathias"}),
+	)
+	_, err := tool.Call(context.Background(), json.RawMessage(
+		`{"owner":"mathias","name":"open-repo","private":false}`,
+	))
 	require.Error(t, err)
 	assert.Contains(t, err.Error(), "confirm")
 }
 
-func TestRepoUpdateTool_MakePublicWithConfirm(t *testing.T) {
-	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
-		w.Header().Set("Content-Type", "application/json")
-		_, _ = w.Write([]byte(`{"name":"infra","full_name":"mathias/infra","default_branch":"main","private":false,"clone_url":"","html_url":""}`))
+// TestRepoUpdateAllowlistRejects: owner outside allowlist denied without network call.
+func TestRepoUpdateAllowlistRejects(t *testing.T) {
+	tool := tools.NewRepoUpdate(
+		gitea.NewClient("http://unused", ""),
+		allowlist.New([]string{"mathias"}),
+	)
+	_, err := tool.Call(context.Background(), json.RawMessage(
+		`{"owner":"evil","name":"some-repo","archived":true}`,
+	))
+	require.Error(t, err)
+}
+
+// TestRepoUpdateUpstreamError: server 500 propagates as ErrUpstream.
+func TestRepoUpdateUpstreamError(t *testing.T) {
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) {
+		w.WriteHeader(http.StatusInternalServerError)
+		_, _ = w.Write([]byte(`{"message":"internal"}`))
 	}))
 	defer srv.Close()
 
-	tool := tools.NewRepoUpdate(gitea.NewClient(srv.URL, "tok"), allowlist.New([]string{"mathias"}))
-	out, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"mathias","name":"infra","private":false,"confirm":"infra"}`))
-	require.NoError(t, err)
-	assert.Contains(t, string(out), `"full_name":"mathias/infra"`)
-}
-
-func TestRepoUpdateAllowlistRejects(t *testing.T) {
-	tool := tools.NewRepoUpdate(gitea.NewClient("http://unused", ""), allowlist.New([]string{"mathias"}))
-	_, err := tool.Call(context.Background(), json.RawMessage(`{"owner":"evil","name":"x"}`))
+	tool := newRepoUpdateTool(srv.URL)
+	_, err := tool.Call(context.Background(), json.RawMessage(
+		`{"owner":"mathias","name":"some-repo","archived":true}`,
+	))
 	require.Error(t, err)
+	assert.ErrorIs(t, err, gitea.ErrUpstream)
 }