NetFabric/NetFabric.Numerics / .agents/skills/copilot-github-actions
copilot-github-actions
UnverifiedAutomate GitHub Copilot CLI in GitHub Actions workflows. USE FOR: installing/running `copilot -p` in workflow steps, choosing GITHUB_TOKEN vs a personal access token (COPILOT_GITHUB_TOKEN) for authentication, setting the copilot-requests write permission, minimal-permission tool allowlisting (--allow-tool, --deny-tool, --no-ask-user, -s/--silent, --output-format=json), triggers (schedule, workflow_dispatch, push), the recommended GitHub Agentic Workflows (gh-aw) alternative to direct CLI invocation, and security/cost considerations (fork PR prompt-injection risk, billing attribution, cost centers). DO NOT USE FOR: embedding the Copilot agent runtime into your own application (use copilot-sdk instead); general GitHub Actions authoring unrelated to Copilot; the Copilot code review or PR features outside of CLI automation.
Copilot CLI + GitHub Actions
Run copilot (the GitHub Copilot CLI) as a step in a GitHub Actions workflow to automate AI-powered tasks in CI/CD — summaries, reports, reviews, scaffolding.
Core concepts
| Concept | Description |
|---|---|
| Workflow pattern | Trigger → checkout/setup → install CLI → authenticate → run copilot -p PROMPT |
| Programmatic mode | -p/--prompt runs one prompt non-interactively and exits; no TTY needed |
GITHUB_TOKEN auth | Built-in Actions token; no secrets to manage; needs copilot-requests: write permission |
| PAT auth | Fine-grained PAT with "Copilot Requests" permission, passed via COPILOT_GITHUB_TOKEN secret |
| Token precedence | COPILOT_GITHUB_TOKEN > GH_TOKEN > GITHUB_TOKEN |
GitHub Agentic Workflows (gh-aw) | Recommended layer over raw CLI invocation; adds guardrails for automated/untrusted contexts |
| Least privilege | Scope permissions:, use --allow-tool/--deny-tool instead of --allow-all/--yolo |
Minimal example (direct CLI invocation)
name: Daily summary
on:
workflow_dispatch:
schedule:
- cron: '30 17 * * *'
permissions:
contents: read
copilot-requests: write
jobs:
daily-summary:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v6
with: { fetch-depth: 0 }
- run: npm install -g @github/copilot
- name: Run Copilot CLI
run: |
copilot -p "Summarize today's commits, with links, under 100 words. Write to summary.md" \
--allow-tool='shell(git:*)' --allow-tool=write --no-ask-user -s
cat summary.md >> "$GITHUB_STEP_SUMMARY"
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
Workflow: automate a task
- Decide direct CLI invocation vs. GitHub Agentic Workflows (recommended for most cases) → gh-aw-vs-direct.md
- Choose auth:
GITHUB_TOKEN(org policy required) or a PAT viaCOPILOT_GITHUB_TOKEN→ authentication.md - Write the prompt/step with minimal permissions and a scoped tool allowlist → cli-invocation-patterns.md
- Review triggers and cost attribution before enabling on fork-PR events or org-wide → security-and-cost.md
Reference Files
| File | Load When |
|---|---|
| references/authentication.md | Setting up GITHUB_TOKEN vs PAT auth, org policy, token precedence |
| references/cli-invocation-patterns.md | Writing the workflow step: prompt options, tool allowlists, output capture, shell patterns |
| references/gh-aw-vs-direct.md | Deciding between raw copilot steps and GitHub Agentic Workflows |
| references/security-and-cost.md | Fork PR risk, prompt injection, least privilege, billing/cost centers |
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 465k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 465k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 465k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 192k | Updated | Claude CodeCodex | Verified | Install |