Skip to content
Back to search

codex-model-routing-team

Route complex Codex tasks to model-specific background workers with bounded concurrency and lead-agent verification.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/zjp1997720--codex-model-routing-team/badge.svg)](/listings/zjp1997720--codex-model-routing-team)

HTML

<a href="/listings/zjp1997720--codex-model-routing-team"><img src="/api/listings/zjp1997720--codex-model-routing-team/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

The standard `skills` CLI shorthand is valid:

npx skills add zjp1997720/codex-model-routing-team

For a global Codex installation without symlinks:

npx skills add zjp1997720/codex-model-routing-team \
  -g -a codex --skill codex-model-routing-team --copy -y

The full GitHub URL works too:

npx skills add https://github.com/zjp1997720/codex-model-routing-team

Verify that the installed package contains both the entrypoint and its supporting policies:

npx skills ls -g -a codex
find ~/.agents/skills/codex-model-routing-team -maxdepth 2 -type f | sort

The file list must include `SKILL.md`, `references/routing-policy.md`, `references/task-packet.md`, and `references/thread-lifecycle.md`. If only `SKILL.md` appears, remove that incomplete installation and install the current release again.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 140 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. > [!IMPORTANT]
  2. > This repository is a generated compatibility mirror. The editable source, Issues, and contributions live in [zjp1997720/zhijian-skills](https://github.com/zjp1997720/zhijian-skills/tree/main/skills/codex-model-routing-team).
  3. # Codex Model Routing Team
  4. <p align="center">
  5. <img src="./assets/readme/hero.svg" width="100%" alt="A lead Codex task routes bounded background work to explicit models">
  6. </p>
  7. <p align="center"><strong>Give Codex a bounded team of model-routed background tasks while one lead keeps control of planning, integration, and verification.</strong></p>
  8. <p align="center"><a href="./README.zh-CN.md">简体中文</a> · <a href="https://github.com/zjp1997720/zhijian-skills/tree/main/skills/codex-model-routing-team">Canonical source</a> · <a href="https://github.com/zjp1997720/codex-model-routing-team">Standalone mirror</a></p>
  9. Use it for complex parallel work when one lead Agent should plan and integrate while bounded background tasks run on explicitly chosen models.
  10. ## Install
  11. The standard `skills` CLI shorthand is valid:
  12. ```bash
  13. npx skills add zjp1997720/codex-model-routing-team
  14. ```
  15. For a global Codex installation without symlinks:
  16. ```bash
  17. npx skills add zjp1997720/codex-model-routing-team \
  18. -g -a codex --skill codex-model-routing-team --copy -y
  19. ```
  20. The full GitHub URL works too:
  21. ```bash
  22. npx skills add https://github.com/zjp1997720/codex-model-routing-team
  23. ```
  24. Verify that the installed package contains both the entrypoint and its supporting policies:
  25. ```bash
  26. npx skills ls -g -a codex
  27. find ~/.agents/skills/codex-model-routing-team -maxdepth 2 -type f | sort
  28. ```
  29. The file list must include `SKILL.md`, `references/routing-policy.md`, `references/task-packet.md`, and `references/thread-lifecycle.md`. If only `SKILL.md` appears, remove that incomplete installation and install the current release again.
  30. ## Activate it
  31. Explicit activation works immediately after installation:
  32. ```text
  33. Use $codex-model-routing-team to research these six independent topics in parallel, then verify and synthesize the findings.
  34. ```
  35. To let Codex activate the Skill automatically for suitable complex work, add the following standing authorization to `~/.codex/AGENTS.md`. Put it in a project-level `AGENTS.md` instead when the authorization should apply only to that project.
  36. ```markdown
  37. ## Codex background model-routing authorization
  38. - The user authorizes Codex to use `$codex-model-routing-team` automatically for complex, parallelizable tasks, create independent background tasks, and assign a model and reasoning level to each task. Before dispatch, briefly state the number of tasks, model, reasoning level, and responsibility. No additional confirmation is required.
  39. - The lead agent keeps its current model and owns planning, file ownership, integration, verification, and final delivery.
  40. - Run at most 6 background tasks concurrently and create at most 8 for one root task. Background tasks must not create more background tasks or subagents.
  41. - Background tasks must not use Ultra. Terra is excluded from automatic routing by default. If Codex App background-task tools are unavailable, complete the work locally and do not use MultiAgentV2 `spawn_agent` as a substitute for model routing.
  42. - Do not auto-dispatch simple questions, status checks, small single-file edits, strongly sequential work, publishing, sending, payment, deletion, account, or production operations.
  43. ```
  44. This is user-configured Codex instruction, not a hidden OpenAI system prompt. Explicit `$codex-model-routing-team` requests remain available without the standing authorization.
  45. ## Why this exists
  46. Codex's native MultiAgentV2 surface does not expose per-worker model or reasoning controls. Native subagents therefore inherit the session model, which can make parallel work unexpectedly expensive.
  47. This Skill uses Codex App background tasks instead. The lead agent plans the work, assigns non-overlapping ownership, verifies results, and integrates the final deliverable. Each background task receives an explicit available model and reasoning level.
  48. ## What it does
  49. - Routes only complex, genuinely parallel work such as multi-source research, multi-section content, large Skills or decks, and independent engineering workstreams.
  50. - Uses Sol and Luna as the default routes, prohibits Ultra, and keeps Terra out of automatic routing unless evidence or the user calls for it.
  51. - Limits fan-out to three new tasks per wave, six concurrent tasks, and eight total tasks per root request.
  52. - Treats the first real task as a health probe, verifies every created task, prevents descendants, and archives only completed tasks whose results were adopted.
  53. - Acts as a Thread Orchestrator for upstream workflows such as Deep Research while preserving their stages, artifacts, and quality gates.
  54. - Keeps publishing, payments, deletion, account changes, and production mutations in the lead task.
  55. ## How it works
  56. 1. The lead agent decides whether parallel execution is worth the coordination cost.
  57. 2. It creates one real background task as a health probe and confirms that the task can be read.
  58. 3. It creates later tasks in bounded waves with explicit model, reasoning, scope, file ownership, and acceptance criteria.
  59. 4. It verifies facts and artifacts, resolves conflicts, and integrates the result.
  60. 5. It archives adopted completed tasks one at a time.
  61. When an upstream Skill already owns decomposition, this Skill accepts its stages and task budget. It controls model routing, task lifecycle, and safety caps without rewriting the upstream workflow. Any task with a workspace output path is project-bound; only chat-only work may be projectless.
  62. The default Deep Research budget is `2-4 researchers + 1 verifier + 1 reviewer + 2 retry slots`, within the cumulative eight-task cap.
  63. ## Example requests
  64. ```text
  65. Use $codex-model-routing-team to implement, test, and review three independent modules without overlapping file ownership.
  66. ```
  67. ```text
  68. Use $codex-model-routing-team to prepare a training deck with separate research, writing, and review tasks.
  69. ```
  70. ```text
  71. Use $codex-model-routing-team as the Thread Orchestrator for $deep-research. Preserve its verifier and reviewer stages.
  72. ```
  73. ## Requirements and boundaries
  74. - Codex App with background-task tools for project discovery, task creation, task reading, follow-up messages, and archiving.
  75. - Access to the models and reasoning levels selected by the lead agent.
  76. - Background task creation must be verifiable. The Skill stops delegation when a task does not materialize.
  77. - This does not change MultiAgentV2 or make native subagents support per-agent model selection.
  78. ## Repository layout
  79. ```text
  80. .
  81. ├── README.md
  82. ├── README.zh-CN.md
  83. ├── LICENSE
  84. ├── skills/
  85. │ └── codex-model-routing-team/
  86. │ ├── SKILL.md
  87. │ ├── agents/
  88. │ ├── evals/
  89. │ └── references/
  90. └── tests/
  91. ```
  92. The agent workflow lives in [SKILL.md](skills/codex-model-routing-team/SKILL.md). Supporting policies live in [references](skills/codex-model-routing-team/references/).
  93. ## Validation
  94. The workflow has been tested with projectless research tasks and project-bound workspace tasks, including model/reasoning verification, result collection, failure handling, and serial archival. The release is also tested through an isolated `npx skills` installation to confirm that supporting files are copied.
  95. ## License
  96. [MIT](LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.