Conversation
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
|
❌ Author of the following commits did not sign a Contributor Agreement: Please, read and sign the above mentioned agreement if you want to contribute to this project |
|
Caution Review failedThe pull request is closed. 📝 WalkthroughWalkthroughThis PR introduces a new "Plan" workflow for GitHub Agent Workflows that enables users to trigger an automated planning assistant via "/plan" commands in issue comments. The workflow generates actionable implementation plans and optionally creates issues or sub-issues to decompose work, alongside supporting documentation and configuration examples. Changes
Sequence DiagramsequenceDiagram
participant User as User<br/>(issue comment)
participant Trigger as trigger-plan.yml<br/>(Trigger Workflow)
participant Main as gh-aw-plan.lock.yml<br/>(Main Workflow)
participant PreActivation as Pre-Activation<br/>(Setup & Context)
participant Agent as Copilot Agent<br/>(Planning)
participant MCP as MCP Gateway<br/>(Tools & Safety)
participant SafeOutputs as Safe Outputs<br/>(Validation)
participant GitHub as GitHub API<br/>(Issues/Comments)
User->>Trigger: Comment with "/plan" on issue
Trigger->>Main: Trigger workflow_call<br/>with COPILOT_GITHUB_TOKEN
Main->>PreActivation: Validate context & checkout
PreActivation->>PreActivation: Generate initial prompt<br/>with repository context
PreActivation->>Main: Return validated context
Main->>Agent: Setup environment<br/>(Go, Python, Node.js, Ruby, uv)
Agent->>Agent: Install Copilot CLI & awf
Agent->>MCP: Initialize MCP gateway<br/>(agents-md-generator,<br/>public-code-search)
Agent->>MCP: Configure GitHub container<br/>integration
Agent->>Agent: Run Copilot agent<br/>with interpolated prompt
Agent->>SafeOutputs: Collect planning output &<br/>validate against schema
SafeOutputs->>SafeOutputs: Process safe outputs<br/>(create-issue, add-comment)
SafeOutputs->>GitHub: Execute validated actions<br/>(create issues/comments)
Main->>Main: Perform threat detection<br/>& parse results
Main->>Main: Upload artifacts<br/>(prompts, logs, threats, manifests)
Main->>GitHub: Post conclusion comment<br/>if needed
Estimated Code Review Effort🎯 4 (Complex) | ⏱️ ~50 minutes Possibly Related PRs
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
There was a problem hiding this comment.
REQUEST_CHANGES: the Plan workflow source is added under github/workflows/ instead of the repository’s canonical .github/workflows/ location, so the corresponding lock workflow is not generated from this PR.
What is this? | From workflow: PR Review
Give us feedback! React with 🚀 if perfect, 👍 if helpful, 👎 if not.
|
@copilot can you please move the workflows to the right place |
|
@strawgate I've opened a new pull request, #330, to work on those changes. Once the pull request is ready, I'll request review from you. |
…330) Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: strawgate <6384545+strawgate@users.noreply.github.com>
Summary
gh-agent-workflows/plan/:example.ymltriggers on issue comments created with a/planprefix (excluding PR comments)README.mddocuments trigger behavior, inputs, and safe outputsgithub/workflows/gh-aw-plan.mdwith:model,additional-instructions,setup-commands,allowed-bot-users,messages-footer)add-comment/create-issuesafe-output usage (including sub-issues viaparent)docs/workflows/gh-agent-workflows.mdmkdocs.ymlValidation
make lint-workflows✅bin/actionlint gh-agent-workflows/plan/example.yml✅make lint-actions✅make docs-buildworkflows/gh-agent-workflows/pr-review-fork.mdlinks to../pr-review/README.md(target not found)Notes
.github/workflows/directly, so the new workflow source was added togithub/workflows/gh-aw-plan.md..github/workflows/gh-aw-plan.mdand runmake compileto generate.lock.ymland dogfood trigger updates.Fixes #319Summary by CodeRabbit
New Features
Documentation