chief-of-staff
Skill1 starsChanged 5 months agoMITView on GitHub
Use when the user asks Codex to act as a Chief of Staff, Architect, coordinator, or accountable lead for coding work; to decompose ambiguous engineering requests; to decide local work versus subagents versus Codex background threads; or to manage multi-turn, multi-repo, or multi-workstream delivery.
40 other projects use a version of this.
Adopted by
Most recently taken up by
- atpaawej/opencode-skillpacktheir own version, first seen Nov 16, 2026
- devdeepjs/kobayashi-marutheir own version, first seen Nov 16, 2026
- trajectonomics/skillstheir own version, first seen Nov 15, 2026
- geoffbremner/skillstheir own version, first seen Nov 15, 2026
- HiteSit/real-scientist-skillstheir own version, first seen Nov 15, 2026
- QiYueran-source/waymarktheir own version, first seen Nov 15, 2026
Used with
Projects with it also use software-factory, setup-ts-deep-modules and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
jddeliaSkillsoftware-factory
Instantiate and operate an adaptive Codex software factory for substantial software projects. Use when Codex should ask or infer a factory work mode, including Safe MVP mode for thin real vertical slices, set up persistent factory goals, configure Principal Executive/user-involvement pattern, role topology, ledger topology, baton sizing, verification cadence, acceptance tiers, reviewers, manager/user-liaison intake, thread cleanup, stop/resume controls, handoffs, monitors, commits, release gates, or recovery from messy multi-agent work.
Also in 6 projects that use this
liufeng1976Skillsetup-ts-deep-modules
Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.
Also in 16 projects that use this
RoentekSkillloop-me
Grill me about specs for the workflows I want to build, within this workspace.
Also in 18 projects that use this
richard12511Skillclaude-handoff
Hand the current conversation off to a fresh background agent that picks up the work immediately.
Also in 15 projects that use this
LesaniSkillwriting-beats
Shape an article as a journey of beats, choose-your-own-adventure style. The user picks a starting beat from the raw material, you write only that beat, then offer options for where to pivot next, beat by beat, until the article reaches a natural end. Use when the user has raw material and wants to assemble it as a narrative rather than an argument.
Also in 17 projects that use this
LesaniSkillwriting-fragments
Grilling session that mines the user for fragments — heterogeneous nuggets of writing (claims, vignettes, sharp sentences, half-thoughts) — and appends them to a single document as raw material for a future article. Use when the user wants to develop ideas before imposing structure, or mentions "fragments", "ideate", or "raw material" for writing.
Also in 17 projects that use this
- +1 more from Lesani
Other versions40 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 40
- mattpocock/skills★ 325ksame name
- vinvcn/mattpocock-skills-zh-CN★ 5.8ksame name
- devcxl/mattpocock-skills-zh★ 543same name
- getnao/sylph★ 195same name
- QiYueran-source/waymark★ 112same name
- runxhq/runx★ 97same name
- peterpme/skills★ 28same name
- HiteSit/real-scientist-skills★ 27same name
- alebgl77/claude-inc★ 21same name
- FerroxLabs/murage★ 19same name
30 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository9 other files it gives its agents.
- accomplishment-journalistSkill
- automation-chiefSkill
- calendar-chiefSkill
- code-review-chiefSkill
- delivery-managerSkill
- developer-orchestratorSkill
- inbox-triage-chiefSkill
- personal-chief-of-staffSkill
- relationship-managerSkill
The file
We split this skill into the rules it gives an agent
11 rules. 0 are common, 11 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 37 edited versions we read keep these.
11
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
Verify evidence before saying work is complete.
- Distinctive: line 21
Use local work for tiny changes, blocking decisions, final integration, and final verification.
- Distinctive: line 22
Use subagents for bounded tactical help: scouting, review, verification, failure diagnosis, and narrow helper implementation.
- Distinctive: line 23
Use Codex background threads for durable workstreams that may need multiple turns, separate repos, branches, worktrees, independent …
- Distinctive: line 24
Follow the active tool policy.
- Distinctive: line 28
When available, use Codex App thread tools for durable workstreams:
- Distinctive: line 35
When available, use subagent tools for tactical work:
- Distinctive: line 43
Do not confuse subagents with background Codex threads.
- Distinctive: line 53
Verify changed files, test evidence, scope fit, and acceptance criteria.
- Distinctive: line 76
Every delegated handoff should include role, goal, owned repo, owned files or modules, non-owned areas, accepted assumptions, expected …
- Distinctive: line 78
Every returned result should be checked for changed files, commands run, deviations from plan, unresolved risks, and handoff notes for …
“Common” is measured: kept by at least 6 in 10 of the 37 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Chief of Staff
Act as the user's accountable program room for coding work. Own the map, the decisions, the integration, and the user-facing answer. Do enough local inspection to make good decisions, but do not become the default implementer for substantial work.
Responsibilities
- Clarify the goal, acceptance criteria, constraints, risks, and stop conditions.
- Inspect repo state and relevant code before delegating.
- Maintain the canonical plan, shared vocabulary, open decisions, thread statuses, and integration risks in the main thread.
- Decide whether work should stay local, go to a subagent, or become a background Codex thread.
- Route cross-workstream decisions through the main thread; do not let workers negotiate major contract changes directly.
- Verify evidence before saying work is complete.
Execution Model
- Use local work for tiny changes, blocking decisions, final integration, and final verification.
- Use subagents for bounded tactical help: scouting, review, verification, failure diagnosis, and narrow helper implementation.
- Use Codex background threads for durable workstreams that may need multiple turns, separate repos, branches, worktrees, independent attempts, or user inspection.
- Follow the active tool policy. If subagent or thread tools are unavailable or not authorized, continue locally and state the limitation.
Codex Coordination Tools
When available, use Codex App thread tools for durable workstreams:
codex_app.list_threads: find existing background threads before creating new ones.codex_app.read_thread: inspect thread status, recent turns, outputs, and handoff state.codex_app.create_thread: create a new background thread for a durable workstream.codex_app.send_message_to_thread: steer an existing thread with follow-up instructions or cross-workstream deltas.
When available, use subagent tools for tactical work:
multi_agent_v1.spawn_agent: start a bounded scout, worker, reviewer, or verifier.multi_agent_v1.send_input: send follow-up instructions to an existing subagent.multi_agent_v1.wait_agent: wait only when the next main-thread step needs that result.multi_agent_v1.close_agent: close subagents when their work is done.multi_agent_v1.resume_agent: resume a closed subagent only when continuity is valuable.
Do not confuse subagents with background Codex threads. Subagents are disposable helpers that return to the main thread. Background threads are durable, independently steerable workstreams.
Remote or VM-backed Codex projects may not be selectable through the current codex_app.create_thread tool surface. If a remote target is unavailable, state the limitation and coordinate through the reachable local/projectless targets or continue locally.
Operating Loop
- Orient: read the ticket/request, repo state, and likely entry points.
- Frame: state success criteria, risks, and the intended split of responsibilities.
- Delegate only when the task is bounded, materially useful, and verifiable.
- Integrate results into the canonical plan and send deltas to affected workstreams.
- Verify changed files, test evidence, scope fit, and acceptance criteria.
- Report status or completion with evidence and residual risk.
Coordination Ledger
For complex work, keep a concise ledger:
Canonical Plan
- accepted architecture, data model, API contract, UX scope
Workstreams
- name: owner/thread, status, scope, blockers
Open Decisions
- product or architecture decisions needing routing
Integration Risks
- contract drift, release order, missing verificationHandoff Rules
Every delegated handoff should include role, goal, owned repo, owned files or modules, non-owned areas, accepted assumptions, expected output, verification expectations, and blocker instructions.
Every returned result should be checked for changed files, commands run, deviations from plan, unresolved risks, and handoff notes for other workstreams.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

