coderabbit-observability
Implements observability for CodeRabbit integrations to track PR coverage, latency, and comment quality.
Install
mkdir -p .claude/skills/coderabbit-observability && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/6309" && unzip -o skill.zip -d .claude/skills/coderabbit-observability && rm skill.zipInstalls to .claude/skills/coderabbit-observability
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Monitor CodeRabbit review effectiveness with metrics, dashboards, andKey capabilities
- →Measure PR review coverage
- →Track comment acceptance rates
- →Calculate time-to-first-review
- →Generate weekly metrics dashboards
- →Alert on missing review coverage
How it works
The skill uses GitHub CLI and API scripts to aggregate PR review data, calculating coverage and comment metrics. It provides templates for GitHub Actions to automate weekly reporting and alerting.
Inputs & outputs
When to use coderabbit-observability
- →Measure PR review coverage
- →Track review latency trends
- →Monitor comment acceptance rates
- →Build team-wide review dashboards
About this skill
CodeRabbit Observability
Overview
Monitor CodeRabbit AI code review effectiveness, review latency, and team adoption. Key metrics include time-to-first-review (how fast CodeRabbit posts after PR creation), comment acceptance rate (comments resolved vs dismissed), review coverage (percentage of PRs reviewed), and per-repository review volume.
Prerequisites
- CodeRabbit installed on GitHub/GitLab organization
- GitHub CLI (
gh) authenticated with org access - Access to CodeRabbit dashboard at app.coderabbit.ai
Key Metrics
| Metric | Target | Why It Matters |
|---|---|---|
| Review coverage | > 90% | PRs without review = blind spots |
| Time-to-review | < 5 min | Fast feedback keeps developers in flow |
| Comment acceptance | > 40% | Low acceptance = noisy reviews |
| Comments per PR | 3-8 | Too many = fatigue, too few = not useful |
| Review state: APPROVED | > 60% | High approval = clean code culture |
Instructions
Step 1: Measure Review Coverage
#!/bin/bash
# coderabbit-coverage.sh - Review coverage for a repo
set -euo pipefail
ORG="${1:?Usage: $0 <org> <repo> [days]}"
REPO="${2:?Usage: $0 <org> <repo> [days]}"
DAYS="${3:-30}"
echo "=== CodeRabbit Review Coverage ==="
echo "Repository: $ORG/$REPO"
echo "Period: Last $DAYS days"
echo ""
TOTAL=0
REVIEWED=0
APPROVED=0
CHANGES_REQUESTED=0
SINCE=$(date -d "$DAYS days ago" +%Y-%m-%dT%H:%M:%SZ 2>/dev/null || date -v-${DAYS}d +%Y-%m-%dT%H:%M:%SZ)
for PR_NUM in $(gh api "repos/$ORG/$REPO/pulls?state=all&per_page=50&sort=created&direction=desc" \
--jq ".[] | select(.created_at > \"$SINCE\") | .number"); do
TOTAL=$((TOTAL + 1))
CR_STATE=$(gh api "repos/$ORG/$REPO/pulls/$PR_NUM/reviews" \
--jq '[.[] | select(.user.login=="coderabbitai[bot]")] | last | .state // "none"' 2>/dev/null || echo "none")
if [ "$CR_STATE" != "none" ] && [ "$CR_STATE" != "null" ]; then
REVIEWED=$((REVIEWED + 1))
[ "$CR_STATE" = "APPROVED" ] && APPROVED=$((APPROVED + 1))
[ "$CR_STATE" = "CHANGES_REQUESTED" ] && CHANGES_REQUESTED=$((CHANGES_REQUESTED + 1))
fi
done
if [ "$TOTAL" -gt 0 ]; then
echo "Total PRs: $TOTAL"
echo "Reviewed by CodeRabbit: $REVIEWED ($(( REVIEWED * 100 / TOTAL ))%)"
echo " Approved: $APPROVED"
echo " Changes Requested: $CHANGES_REQUESTED"
else
echo "No PRs found in the last $DAYS days"
fi
Step 2: Track Comment Volume and Acceptance
set -euo pipefail
ORG="${1:-your-org}"
REPO="${2:-your-repo}"
echo "=== CodeRabbit Comment Analysis ==="
echo ""
TOTAL_COMMENTS=0
PR_COUNT=0
for PR_NUM in $(gh api "repos/$ORG/$REPO/pulls?state=closed&per_page=20" --jq '.[].number'); do
COMMENTS=$(gh api "repos/$ORG/$REPO/pulls/$PR_NUM/comments" \
--jq '[.[] | select(.user.login=="coderabbitai[bot]")] | length' 2>/dev/null || echo "0")
if [ "$COMMENTS" -gt 0 ]; then
TOTAL_COMMENTS=$((TOTAL_COMMENTS + COMMENTS))
PR_COUNT=$((PR_COUNT + 1))
echo "PR #$PR_NUM: $COMMENTS comments"
fi
done
if [ "$PR_COUNT" -gt 0 ]; then
echo ""
echo "Average comments per PR: $(( TOTAL_COMMENTS / PR_COUNT ))"
echo ""
echo "Healthy ranges:"
echo " 1-3 comments/PR → Profile may be too chill"
echo " 3-8 comments/PR → Good signal-to-noise ratio"
echo " 10+ comments/PR → Consider switching to chill profile"
fi
Step 3: Build a GitHub Actions Dashboard
# .github/workflows/coderabbit-metrics.yml
name: CodeRabbit Weekly Metrics
on:
schedule:
- cron: '0 9 * * 1' # Every Monday at 9 AM UTC
workflow_dispatch: # Manual trigger
jobs:
metrics:
runs-on: ubuntu-latest
steps:
- uses: actions/github-script@v7
with:
script: |
const { data: pulls } = await github.rest.pulls.list({
owner: context.repo.owner,
repo: context.repo.repo,
state: 'closed',
per_page: 50,
sort: 'updated',
direction: 'desc',
});
let reviewed = 0;
let approved = 0;
let changesRequested = 0;
let totalComments = 0;
for (const pr of pulls) {
const { data: reviews } = await github.rest.pulls.listReviews({
owner: context.repo.owner,
repo: context.repo.repo,
pull_number: pr.number,
});
const crReview = reviews.find(r => r.user.login === 'coderabbitai[bot]');
if (crReview) {
reviewed++;
if (crReview.state === 'APPROVED') approved++;
if (crReview.state === 'CHANGES_REQUESTED') changesRequested++;
}
const { data: comments } = await github.rest.pulls.listReviewComments({
owner: context.repo.owner,
repo: context.repo.repo,
pull_number: pr.number,
});
totalComments += comments.filter(c => c.user.login === 'coderabbitai[bot]').length;
}
const summary = [
`## CodeRabbit Weekly Metrics`,
`- **Coverage**: ${reviewed}/${pulls.length} PRs reviewed (${Math.round(reviewed/pulls.length*100)}%)`,
`- **Approved**: ${approved}`,
`- **Changes Requested**: ${changesRequested}`,
`- **Avg Comments/PR**: ${reviewed > 0 ? Math.round(totalComments/reviewed) : 0}`,
].join('\n');
core.summary.addRaw(summary).write();
core.info(summary);
Step 4: Set Up Alerts for Review Gaps
# .github/workflows/coderabbit-alert.yml
name: CodeRabbit Review Alert
on:
pull_request:
types: [opened]
jobs:
check-review-expected:
runs-on: ubuntu-latest
steps:
- name: Wait for CodeRabbit review
uses: actions/github-script@v7
with:
script: |
// Wait 10 minutes, then check if CodeRabbit reviewed
await new Promise(r => setTimeout(r, 600000));
const { data: reviews } = await github.rest.pulls.listReviews({
owner: context.repo.owner,
repo: context.repo.repo,
pull_number: context.issue.number,
});
const crReview = reviews.find(r => r.user.login === 'coderabbitai[bot]');
if (!crReview) {
core.warning(
'CodeRabbit has not reviewed this PR after 10 minutes. ' +
'Check: App installation, .coderabbit.yaml, base_branches config.'
);
}
Step 5: CodeRabbit Dashboard Summary
# Build a summary dashboard with these data points:
## Weekly Dashboard Template
| Metric | This Week | Last Week | Trend |
|--------|-----------|-----------|-------|
| PRs opened | | | |
| PRs reviewed by CR | | | |
| Coverage % | | | |
| Avg comments/PR | | | |
| Approval rate | | | |
| Time to first review | | | |
## Action Items:
- Coverage < 90%: Check App installation, base_branches config
- Avg comments > 10: Switch to "chill" profile
- Avg comments < 2: Switch to "assertive" profile
- Approval rate < 50%: Review path_instructions for relevance
Output
- Review coverage metrics calculated per repository
- Comment volume and acceptance rate tracked
- Weekly metrics GitHub Action workflow
- Alert workflow for missing reviews
- Dashboard template for team reporting
Error Handling
| Issue | Cause | Solution |
|---|---|---|
| Coverage below 90% | Some PRs not reviewed | Check base_branches and ignore_title_keywords |
| Low acceptance rate | Too many false positives | Tune path_instructions and switch to chill |
| No metrics data | No closed PRs in period | Extend the time window |
| API rate limited | Too many gh api calls | Add pagination and caching |
Resources
Next Steps
For incident response, see coderabbit-incident-runbook.
When not to use it
- →Managing CodeRabbit account billing
Prerequisites
Limitations
- →API rate limits may affect large repository analysis
- →Requires consistent PR naming and tagging for accurate metrics
How it compares
This method provides quantitative team-wide insights into AI review effectiveness, rather than just individual PR feedback.
Compared to similar skills
coderabbit-observability side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| coderabbit-observability (this skill) | 1 | 27d | Review | Advanced |
| opentelemetry-instrumentation-extension | 3 | 7mo | Review | Advanced |
| phoenix-observability | 3 | 7mo | Review | Intermediate |
| phoenix-tracing | 1 | 29d | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by jeremylongshore
View all by jeremylongshore →You might also like
opentelemetry-instrumentation-extension
docker
Extend OpenTelemetry instrumentation when new functionality is added to the MCP Gateway. Use when (1) new operations/functions are added, (2) reviewing code for missing instrumentation, (3) user requests otel/telemetry additions, or (4) working with state-changing operations. Analyzes git diff, suggests instrumentation points following project standards in docs/telemetry/README.md, implements with approval, writes tests, updates documentation, and verifies with debug logging and docker logs.
phoenix-observability
davila7
Open-source AI observability platform for LLM tracing, evaluation, and monitoring. Use when debugging LLM applications with detailed traces, running evaluations on datasets, or monitoring production AI systems with real-time insights.
phoenix-tracing
Arize-ai
OpenInference semantic conventions and instrumentation for Phoenix AI observability. Use when implementing LLM tracing, creating custom spans, or deploying to production.
sail-voyage
sailresearchco
Use to build or instrument a Sail Voyage — Sail's name for one background or long-running agent run, recorded as a trace of named agents, spans, and events. The entrypoint skill for any Voyage, covering series/version naming, the run→agent→span→event loop, multi-agent structure, running the agent's
agent-performance-monitor
ruvnet
Agent skill for performance-monitor - invoke with $agent-performance-monitor
agent-performance-optimizer
ruvnet
Agent skill for performance-optimizer - invoke with $agent-performance-optimizer