athena-pr-reviewer

name	athena-pr-reviewer
description	PROACTIVELY USED when reviewing a PR, branch, or Jira story. Handles code review against requirements and provides actionable feedback.

Athena PR Reviewer

Instructions

1. Detect PR Target

Parse user input to identify the PR:

• Direct PR reference (PR 123, #123): Extract number directly
• Jira ticket (PROJ-123): Run gh pr list --search "PROJ-123" --json number --jq '.[0].number'
• Current branch: Run gh pr view --json number --jq '.number'
• No PR found: Extract Jira from branch with git branch --show-current | grep -oE '[A-Z]+-[0-9]+'

2. Gather Data (Script)

Run the gather-context script which collects all data in parallel:

~/.claude/skills/athena-pr-reviewer/scripts/gather-context.sh ${PR_NUM} ${JIRA_TICKET}

This script:

• Creates work directory at /tmp/athena-review-${PR_NUM}/
• Fetches in parallel: PR metadata, diff, Jira ticket, epic, CLAUDE.md guidelines, git blame, prior PR comments
• Writes combined context to ${WORK_DIR}/context.md
• Writes diff to ${WORK_DIR}/diff.patch

Output files:

• context.md - Combined PR + Jira + guidelines + history data
• diff.patch - Full PR diff
• pr.json - Raw PR metadata
• jira.json - Raw Jira ticket data
• epic.json - Epic context (if linked)
• guidelines.md - All CLAUDE.md files from repo
• blame.md - Git blame for changed files (who wrote what, when)
• prior-comments.md - Comments from past PRs touching same files

3. Run Reviews (All 8 in Parallel)

Execute all 8 reviews simultaneously in a SINGLE message:

In ONE message, run:

1. Bash (background): Start Gemini + Codex script with run_in_background: true

~/.claude/skills/athena-pr-reviewer/scripts/run-reviews.sh ${WORK_DIR}

2. 6 Task calls: Run all Claude specialists in parallel

This launches all 8 reviews at once:

• Gemini + Codex run in background bash
• 6 Claude agents run as parallel Task calls

After Claude agents complete: Use BashOutput to check if Gemini/Codex finished.

Output files:

• ${WORK_DIR}/reviews/gemini.md
• ${WORK_DIR}/reviews/codex.md
• ${WORK_DIR}/reviews/claude-*.md (6 files)

Claude Specialized Reviews (6 agents)

Each agent reads its prompt file, then analyzes ${WORK_DIR}/context.md and ${WORK_DIR}/diff.patch.

Task: general-purpose
Prompt: "Read ~/.claude/skills/athena-pr-reviewer/prompts/comment-analyzer.md for instructions.
Then read ${WORK_DIR}/context.md and ${WORK_DIR}/diff.patch. Perform the review. Output markdown."
Save to: ${WORK_DIR}/reviews/claude-comments.md

Task: general-purpose
Prompt: "Read ~/.claude/skills/athena-pr-reviewer/prompts/test-analyzer.md for instructions.
Then read ${WORK_DIR}/context.md and ${WORK_DIR}/diff.patch. Perform the review. Output markdown."
Save to: ${WORK_DIR}/reviews/claude-tests.md

Task: general-purpose
Prompt: "Read ~/.claude/skills/athena-pr-reviewer/prompts/error-hunter.md for instructions.
Then read ${WORK_DIR}/context.md and ${WORK_DIR}/diff.patch. Perform the review. Output markdown."
Save to: ${WORK_DIR}/reviews/claude-errors.md

Task: general-purpose
Prompt: "Read ~/.claude/skills/athena-pr-reviewer/prompts/type-reviewer.md for instructions.
Then read ${WORK_DIR}/context.md and ${WORK_DIR}/diff.patch. Perform the review. Output markdown."
Save to: ${WORK_DIR}/reviews/claude-types.md

Task: general-purpose
Prompt: "Read ~/.claude/skills/athena-pr-reviewer/prompts/code-reviewer.md for instructions.
Then read ${WORK_DIR}/context.md and ${WORK_DIR}/diff.patch. Perform the review. Output markdown."
Save to: ${WORK_DIR}/reviews/claude-general.md

Task: general-purpose
Prompt: "Read ~/.claude/skills/athena-pr-reviewer/prompts/simplifier.md for instructions.
Then read ${WORK_DIR}/context.md and ${WORK_DIR}/diff.patch. Perform the review. Output markdown."
Save to: ${WORK_DIR}/reviews/claude-simplify.md

4. Aggregate Reviews

Read all 8 review files and combine findings:

Reviewers:

• Gemini (general)
• Codex (general)
• Claude: comments, tests, errors, types, general, simplify

Confidence Filtering:

• Drop findings with confidence < 80
• Keep findings 50-79 only if flagged by 2+ reviewers

Priority Boost Rule: Items flagged by 2+ reviewers get bumped up one severity level.

Reviewers	Original	Final Severity
3+	High	Critical
2	High	Critical
3+	Medium	High
2	Medium	High
1	Any	No boost

Deduplicate similar findings, noting which reviewer(s) flagged each and average confidence.

4.5 Verify Findings

For each aggregated finding, verify against actual code to filter hallucinations:

1. Read ${WORK_DIR}/diff.patch to get the actual code
2. For each finding with file:line reference:
   • Extract the actual code at that location from the diff
   • Compare the finding's description to what the code actually does
3. Use the verifier prompt (~/.claude/skills/athena-pr-reviewer/prompts/verifier.md) to validate each finding
4. Filter based on verdict:
   • ✓ VERIFIED → Keep in final output
   • ✗ REJECTED → Write to ${WORK_DIR}/rejected.md with reason
   • ⚠️ PARTIAL → Keep but move to "Suggestions" section

Output verified findings to ${WORK_DIR}/verified-findings.md

5. Synthesize Actionable Items

Present combined review to user:

# PR Review: {PR_TITLE} (#{PR_NUM})

## Requirements Status
| Requirement | Status | Notes |
|-------------|--------|-------|

## Action Items (Verified)

### Critical (consensus, verified)
- [ ] file:line - issue - fix [Gemini + Codex + Claude-errors] (3+, avg 92%) ✓

### High Priority (verified)
- [ ] file:line - issue - fix [Gemini + Claude-tests] ← boosted (2, avg 85%) ✓
- [ ] file:line - issue - fix [Claude-types] (95%) ✓

### Medium Priority (verified)
- [ ] file:line - issue - fix [Claude-simplify] (88%) ✓

### Suggestions
- improvements (including PARTIAL findings downgraded from higher severity)

## Rejected Findings
Findings that failed verification are saved to: `${WORK_DIR}/rejected.md`

## Review Sources
- Gemini: ${WORK_DIR}/reviews/gemini.md
- Codex: ${WORK_DIR}/reviews/codex.md
- Claude Comments: ${WORK_DIR}/reviews/claude-comments.md
- Claude Tests: ${WORK_DIR}/reviews/claude-tests.md
- Claude Errors: ${WORK_DIR}/reviews/claude-errors.md
- Claude Types: ${WORK_DIR}/reviews/claude-types.md
- Claude General: ${WORK_DIR}/reviews/claude-general.md
- Claude Simplify: ${WORK_DIR}/reviews/claude-simplify.md

## Recommendation: APPROVE / REQUEST_CHANGES

Examples

User: "Review PR 456"

• Detect PR 456, find linked Jira ticket
• Gather context via script (parallel CLI calls)
• Run 8 reviews in parallel (Gemini, Codex, 6 Claude specialists)
• Aggregate findings, boost items flagged by 2+ reviewers
• Verify findings against actual diff (filter hallucinations)
• Present verified actionable summary

User: "Review CSD-123"

• Find PR linked to CSD-123
• Gather context including acceptance criteria
• 8 parallel reviews (general + specialized)
• Present findings with reviewer attribution

User: "Review this branch"

• Get PR from current branch
• Extract Jira from branch name if needed
• Full 8-reviewer workflow