Skip to content

dayfinggg/openai-codex-agent-skills / codex/skills/orchestration

orchestration

Verified

Plan or coordinate explicitly authorized multi-agent work, including worker briefs, model selection, independent workstreams, integration, and review rounds.

FileKindSize
SKILL.mdskill md1.5 KB
references/delegation-brief.mdmarkdown1.1 KB
references/other-agents.mdmarkdown2.2 KB
references/review-loop.mdmarkdown1.6 KB

references/review-loop.md

View on GitHub
# Multi-ticket implementation and review loops

Use this workflow only when the task explicitly calls for multiple implementation tickets or workers and a tracked integration loop. Ordinary code changes and single reviews do not need a ledger or repeated agent rounds.

## Before dispatch

Record the current base revision and the relevant check result. Write down known failures so that later reports do not attribute them to the new work. Give a reviewer the task text, acceptance conditions, base revision, changed-file boundaries, and the actual diff. Do not include a worker's speculation as a review instruction.

## Ledger

Keep a short ledger in the session scratchpad for multi-step work that must survive handoffs. Record completed tasks, checks, review findings, user decisions, and remaining blockers. Use source files, command output, and revision identifiers as evidence. Do not claim a check passed unless it ran and observed the relevant input.

## Fix rounds

Review the first implementation against every acceptance condition. Give confirmed issues to the owning implementer with the exact finding and a check that would expose it. Review only the follow-up diff and do not repeat findings that were already resolved. If the same issue persists across two focused attempts, stop patching and reassess the underlying approach. Keep scope and approval boundaries unchanged during every round.

## Closing

Integrate only changes that fit the user's request and pass the appropriate combined check. Report unresolved findings or checks honestly. A ledger does not authorize commits, pushes, deployment, or external changes.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall