skill-adaptor

Name: skill-adaptor
Author: zjunlp

Run SkillAdaptor to evolve agent skills from task failures. Use when improving SKILL.md files from trajectories or after benchmark runs.

by zjunlp(0) 0 installs

Free

Free account. Installing gives you the manifest plus copy-paste snippets.

See reviews

About

Imported from zjunlp/SkillAdaptor (plugin/codex/skill-adaptor-plugin/skills/skill-adaptor/SKILL.md). Install upstream with npx skills add zjunlp/SkillAdaptor --skill skill-adaptor. Copyright stays with the author.

SkillAdaptor (Codex)

Evolve reusable SKILL.md files from failure trajectories. Python CLI in this repo.

Quick run

From the SkillAdaptor repo (set skillAdaptorRoot to the directory containing run_plugin.py):

cd <skillAdaptorRoot>
python run_plugin.py init --workspace <workspace> --harness codex
# add task briefs under <workspace>/input_task/*.md
python run_plugin.py --workspace <workspace> --harness codex --max-iterations 2

Outputs: <workspace>/skills/<id>/SKILL.md — auto-synced to ~/.codex/skills/ and <workspace>/.agents/skills/.

Harness

Use --harness codex (not openclaw or claude-code) so adopted skills appear in Codex skill discovery paths.

Docs

See plugin/codex/README.md in the SkillAdaptor repository.

Use it

Copy one of these into your project. Installing also returns the manifest and these snippets.

yaml

targets:
  - https://api.opensmartroute.ai/api/v1/registry/zjunlp-skilladaptor-skill-adaptor/manifest   # or paste the manifest below

Manifest

An Open Capability Manifest: the router reads it to know what this does, what it costs and when to pick it.

zjunlp-skilladaptor-skill-adaptor.ocm.jsonjson

{
  "ocm": "1",
  "id": "zjunlp-skilladaptor-skill-adaptor",
  "kind": "skill",
  "name": "skill-adaptor",
  "description": "Run SkillAdaptor to evolve agent skills from task failures. Use when improving SKILL.md files from trajectories or after benchmark runs.",
  "publisher": "zjunlp",
  "version": "1.0.0",
  "capabilities": {
    "domains": [
      "general"
    ],
    "tags": [
      "skill-md",
      "github"
    ],
    "languages": [
      "en"
    ]
  },
  "quality_prior": 0.6,
  "examples": [
    "Run SkillAdaptor to evolve agent skills from task failures. Use when improving SKILL.md files from trajectories or after benchmark runs."
  ],
  "primary": false,
  "metadata": {
    "source": {
      "provider": "github",
      "repository": "https://github.com/zjunlp/SkillAdaptor",
      "path": "plugin/codex/skill-adaptor-plugin/skills/skill-adaptor/SKILL.md",
      "ref": "b26d1ab5a798f07e53048b5ff509e8535e9fa228",
      "url": "https://github.com/zjunlp/SkillAdaptor/blob/b26d1ab5a798f07e53048b5ff509e8535e9fa228/plugin/codex/skill-adaptor-plugin/skills/skill-adaptor/SKILL.md",
      "key": "zjunlp/SkillAdaptor/plugin/codex/skill-adaptor-plugin/skills/skill-adaptor/SKILL.md"
    }
  },
  "instructions": "# SkillAdaptor (Codex)\n\nEvolve reusable **SKILL.md** files from failure trajectories. Python CLI in this repo.\n\n## Quick run\n\nFrom the SkillAdaptor repo (set `skillAdaptorRoot` to the directory containing `run_plugin.py`):\n\n```bash\ncd <skillAdaptorRoot>\npython run_plugin.py init --workspace <workspace> --harness codex\n# add task briefs under <workspace>/input_task/*.md\npython run_plugin.py --workspace <workspace> --harness codex --max-iterations 2\n```\n\nOutputs: `<workspace>/skills/<id>/SKILL.md` — auto-synced to `~/.codex/skills/` and `<workspace>/.agents/skills/`.\n\n## Harness\n\nUse `--harness ",
  "cost": {
    "context_tokens": 192
  }
}

Fetch it by URL: GET /api/v1/registry/zjunlp-skilladaptor-skill-adaptor/manifest?version=1.0.0