Skip to content
Latchkey

Agent PR Metrics workflow (quay/quay)

The Agent PR Metrics workflow from quay/quay, explained and optimized by Latchkey.

C

CI health: C - fair

Point runs-on at Latchkey and get run de-duplication, job timeouts, self-healing for flaky steps, and up to 58% lower cost, applied automatically.

Grade your own workflow free or run it on Latchkey →
Source: quay/quay.github/workflows/agent-pr-metrics.yamlLicense Apache-2.0View source

What it does

This is the Agent PR Metrics workflow from the quay/quay repository, a real project running GitHub Actions. It is shown here with attribution under its Apache-2.0 license.

Below, Latchkey shows a faster, safer version produced by its optimization engine.

The workflow

workflow (.yml)
name: Agent PR Metrics

on:
  pull_request:
    types: [closed]

permissions:
  contents: read
  pull-requests: write

jobs:
  cyclomatic-complexity:
    name: Cyclomatic Complexity
    if: github.event.pull_request.merged == true && github.event.pull_request.user.login == 'quay-devel'
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
        with:
          persist-credentials: false

      - name: Compute complexity and comment on PR
        env:
          GH_TOKEN: ${{ github.token }}
        run: |
          set -euo pipefail

          PR_NUMBER="${{ github.event.pull_request.number }}"

          ALL_FILES=$(gh api "repos/${{ github.repository }}/pulls/${PR_NUMBER}/files" \
            --paginate --jq '.[].filename')

          # Go: exclude tests, generated protobufs, vendor, config-tool
          CHANGED_GO=$(echo "$ALL_FILES" \
            | grep '\.go$' \
            | grep -v '_test\.go' \
            | grep -v '\.pb\.go' \
            | grep -v '^vendor/' \
            | grep -v '^config-tool/' \
            | xargs -I{} sh -c 'test -f "{}" && echo "{}"' 2>/dev/null || true)

          # Python: exclude test files (test_*.py and *_test.py) and vendor
          CHANGED_PY=$(echo "$ALL_FILES" \
            | grep '\.py$' \
            | grep -v '_test\.py' \
            | grep -v '/test_[^/]*\.py$' \
            | grep -v '^test_[^/]*\.py$' \
            | grep -v '^vendor/' \
            | xargs -I{} sh -c 'test -f "{}" && echo "{}"' 2>/dev/null || true)

          CHANGED=$(printf '%s\n%s' "$CHANGED_GO" "$CHANGED_PY" | grep -v '^$' || true)

          if [ -z "$CHANGED" ]; then
            echo "No Go/Python source files changed - skipping"
            exit 0
          fi

          pip install lizard --quiet

          # lizard --csv columns (no header): NLOC,CCN,token,PARAM,length,location
          LIZARD_OUT=$(echo "$CHANGED" | xargs lizard --csv 2>/dev/null || true)

          if [ -z "$LIZARD_OUT" ]; then
            echo "lizard produced no output - skipping"
            exit 0
          fi

          # Only process data lines (start with a digit)
          MAX_COMPLEXITY=$(echo "$LIZARD_OUT" | awk -F',' '/^[0-9]/{print $2}' | sort -n | tail -1)
          TOTAL=$(echo "$LIZARD_OUT" | awk -F',' '/^[0-9]/{s+=$2} END {print s+0}')
          COUNT=$(echo "$LIZARD_OUT" | awk -F',' '/^[0-9]/{n++} END {print n+0}')
          AVG_COMPLEXITY=$(awk -v total="$TOTAL" -v count="$COUNT" 'BEGIN {
            if (count > 0) printf "%.2f", total / count;
            else print "0"
          }')
          ABOVE_THRESHOLD=$(echo "$LIZARD_OUT" | awk -F',' '/^[0-9]/ && $2 > 15 {n++} END {print n+0}')

          # Build the top-5 most complex functions table
          TOP_FUNCTIONS=$(echo "$LIZARD_OUT" \
            | awk -F',' '/^[0-9]/{print $2","$6}' \
            | sort -t',' -k1 -nr \
            | head -5 \
            | awk -F',' '{printf "| %s | %s |\n", $2, $1}')

          # Build embedded JSON for DevLake ingestion via GitHub connection
          METRICS_JSON=$(jq -cn \
            --argjson max "${MAX_COMPLEXITY:-0}" \
            --argjson avg "${AVG_COMPLEXITY:-0}" \
            --argjson above "$ABOVE_THRESHOLD" \
            --argjson count "$COUNT" \
            '{max_complexity: $max, avg_complexity: $avg, above_threshold: $above, functions_analyzed: $count}')

          # Post as PR comment - DevLake GitHub connection ingests this into
          # pull_request_comments.body; Grafana extracts the agent-metrics JSON.
          gh pr comment "$PR_NUMBER" --repo "${{ github.repository }}" --body "$(cat <<EOF
          ## Cyclomatic Complexity Report

          | Metric | Value |
          |--------|-------|
          | Functions analyzed | ${COUNT} |
          | Max complexity | ${MAX_COMPLEXITY} |
          | Avg complexity | ${AVG_COMPLEXITY} |
          | Above threshold (>15) | ${ABOVE_THRESHOLD} |

          <details>
          <summary>Top 5 most complex functions</summary>

          | Function | CCN |
          |----------|-----|
          ${TOP_FUNCTIONS}

          </details>

          <!-- agent-metrics:${METRICS_JSON} -->
          EOF
          )"

          echo "Posted complexity comment on PR #${PR_NUMBER}: max=${MAX_COMPLEXITY}"

The same workflow, on Latchkey

Removes redundant runs and caps runaway jobs. Added and changed lines are highlighted.

name: Agent PR Metrics
 
on:
  pull_request:
    types: [closed]
 
permissions:
  contents: read
  pull-requests: write
 
concurrency:
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: true
 
jobs:
  cyclomatic-complexity:
    timeout-minutes: 30
    name: Cyclomatic Complexity
    if: github.event.pull_request.merged == true && github.event.pull_request.user.login == 'quay-devel'
    runs-on: latchkey-small
    steps:
      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
        with:
          persist-credentials: false
 
      - name: Compute complexity and comment on PR
        env:
          GH_TOKEN: ${{ github.token }}
        run: |
          set -euo pipefail
 
          PR_NUMBER="${{ github.event.pull_request.number }}"
 
          ALL_FILES=$(gh api "repos/${{ github.repository }}/pulls/${PR_NUMBER}/files" \
            --paginate --jq '.[].filename')
 
          # Go: exclude tests, generated protobufs, vendor, config-tool
          CHANGED_GO=$(echo "$ALL_FILES" \
            | grep '\.go$' \
            | grep -v '_test\.go' \
            | grep -v '\.pb\.go' \
            | grep -v '^vendor/' \
            | grep -v '^config-tool/' \
            | xargs -I{} sh -c 'test -f "{}" && echo "{}"' 2>/dev/null || true)
 
          # Python: exclude test files (test_*.py and *_test.py) and vendor
          CHANGED_PY=$(echo "$ALL_FILES" \
            | grep '\.py$' \
            | grep -v '_test\.py' \
            | grep -v '/test_[^/]*\.py$' \
            | grep -v '^test_[^/]*\.py$' \
            | grep -v '^vendor/' \
            | xargs -I{} sh -c 'test -f "{}" && echo "{}"' 2>/dev/null || true)
 
          CHANGED=$(printf '%s\n%s' "$CHANGED_GO" "$CHANGED_PY" | grep -v '^$' || true)
 
          if [ -z "$CHANGED" ]; then
            echo "No Go/Python source files changed - skipping"
            exit 0
          fi
 
          pip install lizard --quiet
 
          # lizard --csv columns (no header): NLOC,CCN,token,PARAM,length,location
          LIZARD_OUT=$(echo "$CHANGED" | xargs lizard --csv 2>/dev/null || true)
 
          if [ -z "$LIZARD_OUT" ]; then
            echo "lizard produced no output - skipping"
            exit 0
          fi
 
          # Only process data lines (start with a digit)
          MAX_COMPLEXITY=$(echo "$LIZARD_OUT" | awk -F',' '/^[0-9]/{print $2}' | sort -n | tail -1)
          TOTAL=$(echo "$LIZARD_OUT" | awk -F',' '/^[0-9]/{s+=$2} END {print s+0}')
          COUNT=$(echo "$LIZARD_OUT" | awk -F',' '/^[0-9]/{n++} END {print n+0}')
          AVG_COMPLEXITY=$(awk -v total="$TOTAL" -v count="$COUNT" 'BEGIN {
            if (count > 0) printf "%.2f", total / count;
            else print "0"
          }')
          ABOVE_THRESHOLD=$(echo "$LIZARD_OUT" | awk -F',' '/^[0-9]/ && $2 > 15 {n++} END {print n+0}')
 
          # Build the top-5 most complex functions table
          TOP_FUNCTIONS=$(echo "$LIZARD_OUT" \
            | awk -F',' '/^[0-9]/{print $2","$6}' \
            | sort -t',' -k1 -nr \
            | head -5 \
            | awk -F',' '{printf "| %s | %s |\n", $2, $1}')
 
          # Build embedded JSON for DevLake ingestion via GitHub connection
          METRICS_JSON=$(jq -cn \
            --argjson max "${MAX_COMPLEXITY:-0}" \
            --argjson avg "${AVG_COMPLEXITY:-0}" \
            --argjson above "$ABOVE_THRESHOLD" \
            --argjson count "$COUNT" \
            '{max_complexity: $max, avg_complexity: $avg, above_threshold: $above, functions_analyzed: $count}')
 
          # Post as PR comment - DevLake GitHub connection ingests this into
          # pull_request_comments.body; Grafana extracts the agent-metrics JSON.
          gh pr comment "$PR_NUMBER" --repo "${{ github.repository }}" --body "$(cat <<EOF
          ## Cyclomatic Complexity Report
 
          | Metric | Value |
          |--------|-------|
          | Functions analyzed | ${COUNT} |
          | Max complexity | ${MAX_COMPLEXITY} |
          | Avg complexity | ${AVG_COMPLEXITY} |
          | Above threshold (>15) | ${ABOVE_THRESHOLD} |
 
          <details>
          <summary>Top 5 most complex functions</summary>
 
          | Function | CCN |
          |----------|-----|
          ${TOP_FUNCTIONS}
 
          </details>
 
          <!-- agent-metrics:${METRICS_JSON} -->
          EOF
          )"
 
          echo "Posted complexity comment on PR #${PR_NUMBER}: max=${MAX_COMPLEXITY}"
 

What changed

What Latchkey heals here

This workflow has steps that commonly fail on transient issues (network, registries, flaky browsers). On Latchkey managed runners they are detected, retried, and self-healed instead of failing your build:

This workflow runs 1 job per trigger. On Latchkey the same minutes cost up to 58% less than GitHub-hosted, with zero queue time.

Actions used in this workflow