docs: add CLAUDE.md and the conflict-resolver design plan - #10
Conversation
CLAUDE.md distills the repo's enforced rules for coding agents (the one-harness rule, destructive-git-in-scripts-only, the typed exit-code contract, the safety nevers) — apt for a plugin whose whole subject is coding agents. The design plan covers the last big deferred feature, agent-assisted conflict resolution, written plan-first per this repo's own methodology: the journal/resume machinery already makes the resolver a small addition, but the agent-spawn path needs live-herdr probes that are called out explicitly before implementation. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LSffk4LorDDDit18ENESsj
|
Warning Review limit reachedNext included review available in 19 minutes. View limit detailsLimit details: You’ve used the included review currently available. You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. Review configuration: ⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (2)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
🤖 Lab Code Review (draft opinion)The diff adds documentation only; no code changes are present, so no correctness, security, or data-loss risks apply. The documentation is clear and aligns with the repo's conventions. No findings to report. |
2 participants
Two documentation deliverables:
CLAUDE.md— agent onboarding for a repo whose entire subject is coding agents. Distills the rules the test suite actually enforces (one shared harness, destructive-git-in-scripts-only, the lockstep exit-code contract, the herdr-wrapper invariant, the safety nevers) plus commands and docs conventions, so any agent dropped into this repo works inside its guardrails from turn one.docs/plans/2026-08-24-002-feat-conflict-resolver-agent-plan.md— the design for the last big deferred feature, agent-assisted conflict resolution, written plan-first per this repo's own methodology. Key insight: the journal/resume machinery makes the resolver small — it sits exactly where the human[s]hell into merge treesits, and everything downstream (resume_offer swap confirmation, abort-merge escape hatch) already exists. The plan restricts spawning to the plugin-owned detached-locus merge tree (user-tree conflicts refused, per the Locus KTD), keeps the swap behind the existing explicit confirmation, and calls out the three live-herdr probes needed before implementation.No code changes; 237 tests green.
🤖 Generated with Claude Code
https://claude.ai/code/session_01LSffk4LorDDDit18ENESsj
Generated by Claude Code