Adds extensive skill documentation for GitHub Agentic Workflows (2026 features) covering AI-powered repository automation, MCP integration, security architecture, and operational patterns.

Skill Created

.github/skills/github-agentic-workflows/SKILL.md (1670 lines, 34KB)

Coverage

13 Operational Patterns:

• ChatOps (slash commands), DailyOps (scheduled improvements), DataOps (extraction + analysis)
• IssueOps (triage), LabelOps (label triggers), MemoryOps (stateful workflows)
• MultiRepoOps (cross-repo), ProjectOps (board automation), TaskOps (scaffolded improvements)
• SideRepoOps, SpecOps, DispatchOps, TrialOps

Security Architecture (Defense-in-Depth):

• 6 layers: compile-time validation, runtime isolation, permission separation, network controls, output sanitization, safe actions
• Safe inputs/outputs (7 types: issue, comment, PR, project, asset, SARIF, minimize)
• Threat detection, prompt injection protection, SARIF reporting

Tools & Integration:

• MCP servers (GitHub, filesystem, web, Playwright, bash, custom)
• MCP Gateway for unified HTTP access
• Memory systems: cache-memory (7-day), repo-memory (unlimited Git branch storage)
• Orchestration: orchestrator/worker patterns with tracker IDs

Examples:

• 5 production-ready recipes (issue triage bot, daily code quality improver, PR review assistant, weekly status report, security scan coordinator)
• CLI commands reference
• Cost management strategies
• Security best practices (10 DO, 8 DON'T guidelines)

Documentation Updates

• SKILLS.md: 40 → 41 skills, added comprehensive description
• README.md: Updated skill count, expanded Development & Operations category (10 → 11)

Source Analysis

Analyzed official GitHub Agentic Workflows documentation:

• gh-aw/setup/creating-workflows/
• gh-aw/introduction/how-they-work/
• gh-aw/reference/glossary/
• Model Context Protocol specs

All 2026 features included: repo-memory, tracker IDs, threat detection, safe output messages, GitHub Projects v2 integration, asset uploads, comment minimization.

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.