Add Claude-led orchestration skill (Plan → Tasks → Execute → Review) with docs and contract test#1
Merged
Conversation
Copilot
AI
changed the title
[WIP] Implement orchestration flow for Claude and Codex
Add Claude-led orchestration skill (Plan → Tasks → Execute → Review) with docs and contract test
Jun 12, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR adds an explicit orchestration contract for the main Claude thread to run end-to-end workflow in the intended “Claude plans/reviews, Codex executes” model. It fills the current gap between existing
/codex:*commands and internal single-shot skills while preservingcodex:codex-rescueas a forward-only subagent.New orchestration skill
plugins/codex/skills/codex-orchestration/SKILL.mdas a user-invocable skill.PLAN: architecture decisions, target files, constraints, edge cases, implementation brief.TASKS: small bounded tasks with explicit acceptance criteria.EXECUTE: delegate via/codex:rescue(prefer--backgroundfor long work), monitor with/codex:status, collect via/codex:result, continue with--resume.REVIEW: validate against brief, optionally use/codex:reviewor/codex:adversarial-review, loop back through/codex:rescue --resumeuntil passing.README flow documentation
Orchestration: Plan → Tasks → Execute → Review.Test coverage for orchestration contract
tests/orchestration.test.mjs.name,user-invocable) is correct,codex-rescueremains single-shot forwarder).Example contract excerpt: