tune-worker-routing
Use when reviewed external-worker outcomes should update local task-routing recommendations or support a proposed stable-policy change.
Skill by MAbbara
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/mabbara-tune-worker-routing
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/tune-worker-routing
What's inside
- Skill name
tune-worker-routing- Description
- Use when reviewed external-worker outcomes should update local task-routing recommendations or support a proposed stable-policy change.
- License
- Apache-2.0
- Instructions
- 13 lines in SKILL.md · view on GitHub
Files (3)
- SKILL.md
- openai.yaml · agents/
- profile-fields.md · references/
SKILL.md (first lines)
# Tune Worker Routing Resolve `WORKER_CLI` from this installed skill directory as `../delegate-kimi-k3/scripts/worker-cli.mjs`, then convert it to an absolute path. Run worker commands as `node "$WORKER_CLI" ...`. If the file is missing, stop and report a broken plugin installation; never download a replacement. 1. Validate each reviewed outcome and keep patch acceptance separate from task fit. 2. Append to `.external-worker/review-ledger.jsonl`; never overwrite prior evidence. 3. Run `node "$WORKER_CLI" profile --repo <repo> --regenerate`. 4. Treat 1–2 comparable runs as evidence only, 3–4 as tentative, and 5+ as eligible for generated routing changes only when the conservative confidence bound supports it. 5. Segment model/CLI changes; prioritize explicit human acceptance and task-fit feedback over model scores. Do not let one run reverse policy. 6. Never enable prohibited ownership or weaken review/human gates. Corrupt profiles fall back to safe defaults. 7. When stable instructions should change, create a provenance-rich proposal citing run IDs, old/new behavior, benefits, regressions, and a patch. Run policy tests and require explicit human application. Never let Kimi approve its own delegation skill.
From MAbbara/codex-kimi-worker, Apache-2.0.
More from MAbbara/codex-kimi-worker
- 0.479score
delegate-kimi-k3
MAbbara
Use when considering Kimi K3 for a bounded software-engineering task, including UI, 3D, reconnaissance, tests, focused fixes, or measured experiments.
SkillCFreshness: CClaude Code +9 - 0.474score
review-external-worker-run
MAbbara
Use when an external-worker evidence bundle needs independent Codex review, adjudication, repair routing, or final acceptance assessment.
SkillCFreshness: CClaude Code +9 - 0.473score
setup-review-agents
MAbbara
Use when installing, inspecting, or uninstalling codex-external-worker reviewer-agent templates for a project.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jul 16, 2026 (3 months ago)
- Last release
- Jul 16, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/MAbbara/codex-kimi-worker
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from MAbbara
- 0.479score
delegate-kimi-k3
MAbbara
Use when considering Kimi K3 for a bounded software-engineering task, including UI, 3D, reconnaissance, tests, focused fixes, or measured experiments.
SkillCFreshness: CClaude Code +9 - 0.474score
review-external-worker-run
MAbbara
Use when an external-worker evidence bundle needs independent Codex review, adjudication, repair routing, or final acceptance assessment.
SkillCFreshness: CClaude Code +9 - 0.473score
setup-review-agents
MAbbara
Use when installing, inspecting, or uninstalling codex-external-worker reviewer-agent templates for a project.
SkillCFreshness: CClaude Code +9
Cite this page
tune-worker-routing skill. Agent Mart. https://agentmart.dev/skills/mabbara-tune-worker-routing (last updated 2026-10-24)