Add Q agentic workflow for optimizing workflows

[Copilot]

• Understand the requirements for the Q workflow (James Bond reference)
• Create the workflow file .github/workflows/q.md with all features
• Compile the workflow to generate .lock.yml file (235KB)
• Test the workflow configuration
• Run make agent-finish for final validation (all passed)
• Code review completed
• Address PR feedback:
  • Wrapped Current Context section in XML tags (<current_context>) for better prompt parsing
  • Added explicit mention of safe-outputs MCP server in Phase 6
  • Verified strict mode is already enabled (strict: true)
  • Recompiled workflow successfully (235.8 KB)
  • All tests passing

Original prompt

Add an agentic workflow, q (récente de James bond) whose task is to improve/optimize/fix Agentic workflows.

This workflows starts with the command "/q" on an issue or an issue comment and processes the current text context. Creates a pull request with changes.

Imports the gh-aw,serena,Tavily shared workflows.

It should ignore changes to lock files and let copilot agent compile them later.

It should use the gh-aw tools to investigate logs and audits to get live information from agentic workflows. NEVER EVER make up a log, always pull from live data.

Use the gh-aw compiler to ensure changes to the workflows are syntactically correct. Keep changes to a minimum.

Areas to look into: adding missing tools (from missing tools report, permissions issue), detecting poor performance through excessive repetitive mcp cools, try to extract common execution pathway and generate those as steps in the workflow.

Uses copilot engine.

---

💡 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.

[pelikhan]

@copilot

• mention safe outputs
• use compilation in strict mode

[pelikhan]

@copilot wrap this content in xml tags following prompt best practices

[Copilot]

Wrapped the Current Context section in <current_context> XML tags following prompt best practices. This makes it easier for the AI to parse structured information. (commit b164bbd → updated)

[Copilot]

@copilot

• mention safe outputs...

✅ Both items addressed:

1. safe-outputs: Added explicit mention of the safe-outputs MCP server in Phase 6, explaining to use the create-pull-request tool from the safe-outputs MCP server
2. strict mode: Already enabled in frontmatter (strict: true on line 23)

Workflow recompiled successfully (235.8 KB). (commit b164bbd → updated)

[github-actions]

Agentic Changeset Generator triggered by this pull request