feat(self-driving): ground custom-scout gap analysis in for-agents context#876
Merged
Conversation
…ntext The 6b gap analysis now starts from the repo's agent-facing docs (AGENTS.md, CLAUDE.md, ARCHITECTURE.md, .cursor/rules) when present, instead of going straight to scanning source. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
🧙 Wizard CIRun the Wizard CI and test your changes against wizard-workbench example apps by replying with a GitHub comment using one of the following commands: Test all apps:
Test all apps in a directory:
Test an individual app:
Show more apps
Results will be posted here when complete. |
Contributor
There was a problem hiding this comment.
Pull request overview
Updates the self-driving program’s STEP 6b guidance so custom-scout gap analysis is grounded in repo-maintained “for-agents” context (e.g., AGENTS/CLAUDE/ARCHITECTURE/Cursor rules) before scanning source, improving the relevance and specificity of proposed scouts.
Changes:
- Adjusted STEP 6b prompt text to instruct the agent to start from “for-agents” context when present.
- Updated the self-driving ARCHITECTURE step summary and step table to reflect the same guidance.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
| src/lib/programs/self-driving/prompt.ts | Refines STEP 6b prompt wording to ground custom-scout gap analysis in agent-facing repo context first. |
| src/lib/programs/self-driving/ARCHITECTURE.md | Aligns documentation for STEP 6b (summary + table) with the updated prompt behavior. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
sortafreel
approved these changes
Jul 13, 2026
sortafreel
reviewed
Jul 13, 2026
Contributor
Author
|
skill-side mirror (source of truth for the HOW): PostHog/context-mill#233 — merge that one for the change to take effect |
dkarasiewicz
added a commit
to PostHog/context-mill
that referenced
this pull request
Jul 14, 2026
…ntext (#233) Mirrors PostHog/wizard#876 — the skill references are the source of truth for the HOW, so the wizard prompt change alone doesn't take effect. Co-authored-by: Dawid Karasiewicz <dawid.k@posthog.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
dkarasiewicz
enabled auto-merge (squash)
July 14, 2026 14:40
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
step 6b custom scout proposals go straight to scanning source and skip the repo's agent-facing docs (AGENTS.md, CLAUDE.md, ARCHITECTURE.md, .cursor/rules). those already map the product's surfaces, so proposals come out slow and generic.
Changes
Test plan
pnpm build + self-driving prompt unit test pass, covered by CI
LLM context
authored with claude code, commit created via createCommitOnBranch to satisfy the signed-commits rule
🤖 Generated with Claude Code