update spec workflow to use repo-canonical two-layer model#1865
update spec workflow to use repo-canonical two-layer model#1865transphorm merged 2 commits intodevfrom
Conversation
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
|
Caution Review failedThe pull request is closed. ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (7)
📝 WalkthroughWalkthroughUpdated spec/planning workflow: repo-based canonical markdown execution specs are created under Changes
Sequence Diagram(s)sequenceDiagram
participant Skill as .claude/spec-from-audit (skill)
participant Repo as Git Repository
participant LinearAPI as Linear API
participant Issue as Linear Issue
Skill->>Repo: create/update `specs/.../plans/<ID>-<slug>.md`
Note right of Repo: canonical, version-controlled spec
Skill->>LinearAPI: create Linear document (mcp__linear-server__create_document)
LinearAPI->>Issue: attach document to Issue (issue parameter)
LinearAPI-->>Skill: return document link/ID
Skill->>Repo: update workstream `SPEC.md` backlog entry to reference plan
Repo-->>Skill: confirm commit/push
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Suggested labels
Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 inconclusive)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.claude/skills/spec-from-audit/SKILL.md:
- Around line 38-42: Change the process so updating the workstream backlog row
is mandatory for every generated plan: when writing a plan to
specs/projects/sdk/workstreams/<scope>/plans/<ID>-<slug>.md via the Write tool,
always also create or update the corresponding workstreams/<scope>/SPEC.md
backlog row (do not use “if one exists” conditional); likewise, still create the
Linear document using mcp__linear-server__create_document linked to the issue
(not the project) so both repo file and cross-tool copy are produced and the
workstream SPEC.md always reflects the new backlog entry.
In `@CLAUDE.md`:
- Line 65: Resolve the PR size policy conflict by making the threshold in the
"5. **Then implement** — one spec = one PR ≤2k LOC" line consistent with the
"Key Rules" section; either replace "≤2k LOC" with the "1k–3k LOC" range used in
Key Rules or remove the specific numeric limit from the "Then implement" bullet
so it defers to Key Rules as the single authoritative threshold.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: e5c4b580-322d-4df2-8b29-045d78783eb0
📒 Files selected for processing (4)
.claude/skills/spec-from-audit/SKILL.mdCLAUDE.mdapp/AGENTS.mdpackages/mobile-sdk-alpha/AGENTS.md
Summary
specs/) are the canonical source and Linear documents are mirrored copies for cross-tool accessspecs/folder was marked deprecated); this reverses that decision so specs are version-controlled and agent-executable from the repo/spec-from-auditskill to write specs to both repo and LinearChanges
Docs/specs
CLAUDE.md— updated Specs & Planning section: specs live in both repo and Linear, repo is canonical; restored two-layer model (backlog inSPEC.md, plans inplans/)app/AGENTS.md— point implementation specs to repospecs/as canonical sourcepackages/mobile-sdk-alpha/AGENTS.md— same updateConfig/infra
.claude/skills/spec-from-audit/SKILL.md— updated skill to write specs to repo first, then mirror to Linear documentTest Plan
🤖 Generated with Claude Code
Summary by CodeRabbit