Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/igovet/codex-cortex-orchestrator/documentation-syncnpx skills add igovet/codex-cortex-orchestrator --skill documentation-syncgit clone --depth 1 https://github.com/igovet/codex-cortex-orchestratorWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00027 | $0.00840 |
| Opus 5 | $0.00014 | $0.00420 |
| Sonnet 5 | $0.00005 | $0.00168 |
| Haiku 4.5 | $0.00003 | $0.00084 |
Grade A, and why
documentation-sync scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Documentation Sync
Update only durable knowledge changed by completed, verified work.
The coordinator must not read the target diff, source, tests, or documentation, and must not edit or verify documentation directly. It routes existing report evidence and delegates every missing inspection, technical judgment, edit, command, artifact/state check, and verification action. This includes proving that expected documentation or project-local artifacts are absent or unchanged.
After project verification, the coordinator makes the documentation-impact decision from its bounded index-driven knowledge-route context and verified implementation and verification reports. This is a model-owned outcome obligation before closure/final synthesis, not a backend stage, gate, or permission. Consider behavior, architecture, interfaces, commands, verification, conventions, feature ownership, public usage, and operating expectations. Do not inspect additional project files to fill an evidence gap.
When durable knowledge changed:
- If a required index is missing or unreadable, or the routed knowledge and reports do not identify affected documentation, create a bounded discovery delegation. A reported stale, conflicting, or incomplete page receives follow-up only when its task impact warrants it; do not force a harvest. The coordinator does not inspect source or additional documentation itself.
- Create a dedicated documentation-sync worker, normally using the advisory
technical_writerprofile. Compose the orchestrator's authoritative delegation knowledge contract into its instructions with the relevant finalized evidence references and a concise textual scope boundary. - Assign every affected surface among the harvest documents in
docs/project/anddocs/features/plus affected public documentation. Do not edit unrelated pages simply to demonstrate that synchronization ran. - Instruct the writer to preserve manual text and update only generated blocks unless the task explicitly authorizes a manual-document change.
- For material documentation impact, create a separate verification delegation. That worker verifies documentation against current source, tests, commands, links, feature ownership, and the change reports, then publishes the evidence through the applicable semantic documentation publication operation.
- Use writer and verification reports to decide whether corrective documentation work or a residual-risk disclosure is needed.
- When the active tool returns a current contained digest-verified final report or plan Markdown projection, give the user its clickable absolute path with a localized summary of what changed and verification status. Never publish a bare, guessed, or stale path. Task, decision, delegation, initiative, closure, governance, handoff, index, and timeline records are SQLite-only and have no user-facing Markdown path. Projection failure is a human-view limitation, not evidence that documentation work failed.
For the documentation not required path, when the reports establish that no
durable documentation surface changed, do not create a writer merely to record
that the task occurred. Require one
finalized worker-owned report with an explicit English documentation-impact
section and material/no-impact rationale. An existing implementation or
verification report qualifies only when it already contains that explicit
section. When the evidence is absent or spread across reports, the coordinator
creates a bounded evidence-synthesis/documentation-impact assignment, passes
the exact finalized evidence references, waits, and reads the worker-submitted
finalized rationale publication; the coordinator never publishes on behalf of
a worker or self-asserts documentation_not_required. The orchestrator links
the exact task, that documentation-impact evidence, and every other required
finalized evidence in the final initiative, then cites the server-returned
references and digests in the closure evidence
evidence before task-scoped and initiative-scoped governance inspection.
For light/full governance, missing or unconsumed required documentation-impact
records reject closure, while still never requiring a documentation edit for a
no-impact report and never blocking an honest user-facing explanation of the
limitation.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 74 lines · 27 tokens per session scan A c73308bd3862
documentation-sync is a skill published in the GitHub repository igovet/codex-cortex-orchestrator (1 stars, last pushed 2d ago), licensed MIT. It adds 27 tokens to every session and 840 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
bug-triage
Triage bugs reported in chat/issues, search for duplicates, file or update GitHub issues with full context, and push fix PRs.
ao-desktop-dev
Launch, restart, or troubleshoot the real AO Electron desktop app from this repository; run a checkout against isolated or real local AO data; combine PR branches for local UI review; and diagnose stale Electron processes, port conflicts, or preload bridge mismatches. Use whenever asked to run, open, show, or visually…
using-ao
Catalog of the AO (Agent Orchestrator) ao CLI: spawning workers, managing sessions and projects, sending messages, controlling the shared browser, previewing pages, and daemon control. Use when using the ao CLI, spawning workers, or managing AO sessions in an AO workspace.
model-router
当需要调度不同模型执行子任务时使用。根据任务类型匹配模型能力,自动选择最优模型并生成调用指令。TRIGGER: 模型调度、model dispatch、选模型、分派任务、多模型协作、路由到模型.
goal-flight
Portable Goal Flight workflow for long-running repo work: planning, dispatch, review, recovery, file-backed resume.
agent-orchestrator-v2
Agent Orchestrator workflow skill. Use this skill when the user needs Meta-skill que orquestra todos os agentes do ecossistema. Scan automatico de skills, match por capacidades, coordenacao de workflows multi-skill e registry management and the operator should preserve the upstream workflow, copied support files, and…