Agent skills for Devin Desktop

Monitor long-running commands, asynchronous artifacts, callbacks, dispatched Codex work, and Slurm jobs with detached deterministic…

Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.

Orchestrate implementation, orchestration complete: route work among the Codex primary task, disposable in-thread subagents, and…

Use for natural-language questions or requests about whether Kimi K3 or another audited External Model is available or callable as…

Send a design document to Codex CLI for review, then revise the document based on the findings.

Update the title and body of one or more pull requests.

QA the omo Codex Light edition (lazycodex / packages/omo-codex) itself, in strict isolation so ONLY our plugin is exercised, never the…

Diagnose and safely repair Codex "Reconnecting... 1/5" through "5/5" loops in the CLI, desktop app, or IDE on macOS, Linux, and Windows.

Internal guidance for presenting Codex helper output back to the user

Review documents using Codex exec. Supports review loop with context preservation.

Quick second-opinion using Codex MCP (diff only, no tests).

Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment…

Internal helper contract for calling the codex-companion runtime from Claude Code

Second-opinion review of the current diff via the Codex CLI, read-only.

Run OpenAI Codex via the Codex App Server (JSON-RPC-over-stdio) for streaming multi-turn chat sessions with persistent threads, structured…

Audit local Codex session health by engagement recency and safely cold-store or restore older top-level sessions without deleting history.

Orchestrates multi-session agent workflows, subagent dispatching, QA verification, and repair gates across complex development lifecycles.

Audits, streamlines, and refactors Codex and AI agent skills to eliminate redundancy and optimize prompt overhead.

Coordinate multiple Codex agents.

Delegate a bounded analysis, implementation proposal, or independent review to a locally running Worker Gateway, then critically inspect…

Delegate coding tasks to Codex, Claude Code, or Pi agents via background process.