Subagent Orchestrator
Shen-Hua orchestration workflow for coordinating multiple Codex subagents or simulated specialist roles. Use when the user explicitly asks for subagents, parallel agents, delegation, or multi-agent work, or when a complex task can be split into independent explorer, architect, worker, tester, reviewer, or docs-researcher streams under the current runtime policy.
Handoff Writer
Shen-Hua handoff writer for ending or pausing long-running Codex work. Use at the end of every project work round, before final responses, after interruptions, when switching agents/sessions, or whenever the next Codex instance needs a concise resume point with completed work, changed files, verification, risks, and next actions.
Self Reviewer
Shen-Hua self-review workflow before finishing a task. Use before final responses, commits, PRs, handoffs, or readiness claims to inspect requirements coverage, diffs, risks, tests, missed edge cases, destructive changes, and unresolved work.
Verification Loop
Shen-Hua verification and repair loop for modified projects. Use after code/config/docs changes that need quality gates, before final readiness claims, when tests/build/lint/typecheck fail, before a PR/commit, or when the user asks Codex to check, verify, test, fix failures, or prove the current work is safe.
Shen Hua
Top-level Shen-Hua long-task agent entrypoint. Use when the user invokes `$shen-hua` or asks Codex to run the full Shen-Hua workflow for complex project work, including planning, runtime queueing, memory retrieval, subagent orchestration, verification, self-review, handoff, automation preparation, and external-sync preparation.
Project Commander
Shen-Hua long-task commander for complex project work. Use when the user asks Codex to complete a project, coordinate a multi-step goal, continue long-running work, recover from an interrupted task, manage several skills/subagents, or turn a broad objective into an executable plan with task tracking, memory updates, verification, review, and handoff.
Deep Planner
Shen-Hua deep planning workflow for complex tasks before implementation. Use before large edits, refactors, architecture changes, public API changes, cross-module work, risky migrations, unclear requirements, or when the user asks Codex to investigate first, produce a deep plan, identify risks, or plan a long-running implementation.
Persistent Memory
Shen-Hua project memory manager. Use whenever Codex needs to initialize, read, repair, update, or summarize the project's `.agent/` working memory files, including PROJECT_BRIEF.md, PLAN.md, TASKS.md, LOG.md, DECISIONS.md, HANDOFF.md, and REVIEW_CHECKLIST.md for long-running tasks.
Runtime Operator
Shen-Hua V2 runtime operator for minimizing the gap to model-level long-task agents. Use when Codex needs real task queueing, durable event logs, memory indexing/search, resume prompt generation, verification gates, automation handoff prompts, or external-sync outbox files for long-running project work.