fix(review): give Blind Hunter project access; verbatim skill prompts#2507
Conversation
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (3)
📝 WalkthroughWalkthroughUpdated three review-step documents so Blind Hunter and Edge Case Hunter are invoked with explicit skill prompts on ChangesReview prompt wording updates
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~4 minutes Possibly related PRs
Suggested reviewers
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 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 |
🤖 Augment PR SummarySummary: Rewrites the Blind Hunter and Edge Case Hunter reviewer instructions to be explicit, verbatim prompts that apply the corresponding review skills to 🤖 Was this summary useful? React with 👍 or 👎 |
The Blind Hunter review subagent was denied project access in dev-auto,
quick-dev, and code-review — limited to the diff, unable to inspect the
surrounding code that determines whether the change is correct. It is blind
to intent (no spec/context), not to the codebase.
Replace the prose reviewer descriptions with bare verbatim prompts that
invoke the named review skill on {diff_output}. Subagents already have
project read access by default, so the explicit prompt is the only control
needed; this also removes the vague "invoke via the skill" phrasing that
invited emulating the review instead of actually invoking it.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
alexeyv
force-pushed
the
fix/blind-hunter-project-access
branch
from
14a788b to
e6f1595
Compare
![augmentcode[bot]](https://avatars.githubusercontent.com/in/1027498?s=60&v=4){kind=small}
|
|
||
| - **Blind hunter** — receives inline `{diff_output}` only. No spec, no context docs, no project access. Invoke via the `bmad-review-adversarial-general` skill. | ||
| - **Edge case hunter** — receives `{diff_output}` and read access to the project. Invoke via the `bmad-review-edge-case-hunter` skill. | ||
| - **Blind hunter** — prompt: "Use the `bmad-review-adversarial-general` skill on `{diff_output}`." |
There was a problem hiding this comment.
This references another skill but uses “Use … skill” phrasing; tools/skill-validator.md REF-03 requires “invoke” language (e.g., “Invoke/Launch the … skill”) when naming a skill. Consider aligning the wording to avoid failing skill validation. (Guideline: skill_validation) (Rule: tools/skill-validator.md)
Severity: high
Other Locations
src/bmm-skills/4-implementation/bmad-dev-auto/step-04-review.md:29src/bmm-skills/4-implementation/bmad-quick-dev/step-04-review.md:27src/bmm-skills/4-implementation/bmad-quick-dev/step-04-review.md:28src/bmm-skills/4-implementation/bmad-code-review/steps/step-02-review.md:21src/bmm-skills/4-implementation/bmad-code-review/steps/step-02-review.md:23
🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.
| 2. Launch parallel subagents without conversation context. If subagents are not available, generate prompt files in `{implementation_artifacts}` — one per reviewer role below — and HALT. Ask the user to run each in a separate session (ideally a different LLM) and paste back the findings. When findings are pasted, resume from this point and proceed to step 3. | ||
|
|
||
| - **Blind Hunter** — receives inline `{diff_output}` only. No spec, no context docs, no project access. Invoke via the `bmad-review-adversarial-general` skill. | ||
| - **Blind Hunter** — prompt: "Use the `bmad-review-adversarial-general` skill on `{diff_output}`." |
There was a problem hiding this comment.
The RULES section above still says Blind Hunter “receives NO project context — diff only,” which seems inconsistent with this PR’s goal of fixing the “no project access” behavior. If that RULES text is used to decide subagent permissions, the bug may persist despite the new prompt line.
Severity: medium
🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.
1 participant
What
Fixes the Blind Hunter review subagent across
bmad-dev-auto,bmad-quick-dev, andbmad-code-review, and makes the review-skill invocation explicit.Why — the bug
All three skills described the Blind Hunter as receiving the diff with "no project access." That's wrong. "Blind" means blind to intent — no spec, no context docs, so it can't anchor on why the change was made — not blind to the codebase. Denying it project access crippled it: it could only see the diff, never the surrounding code that determines whether the change is actually correct.
The change
Replace the prose reviewer descriptions with bare, verbatim prompts:
…skill" prompt also replaces the vague "invoke via the skill" phrasing that previously left room to emulate the review instead of actually invoking the skill.bmad-code-review's Acceptance Auditor is unchanged — it's the spec-aware reviewer by design.Scope
Three one-line-per-reviewer edits;
npm run qualitypasses. Separate from #2506 (dev-auto commit-at-end); the two PRs touchdev-auto/step-04-review.mdin different hunks.🤖 Generated with Claude Code