subagents
Skill1 starsChanged 5 days agoMITView on GitHub
Delegate main or independent subtasks to subagents.
12 other projects use a version of this.
Adopted by
Most recently taken up by
- haxqer/skillstheir own version, first seen Oct 20, 2026
- ershiyidian/QuayMeshtheir own version, first seen Oct 15, 2026
- nimadorostkar/Claude-Skills-collectiontheir own version, first seen Oct 13, 2026
- sebnado/worldbuild-benchtheir own version, first seen Oct 11, 2026
- wtsi-hgi/agentskillstheir own version, first seen Oct 4, 2026
- nextacademy-io/mastering-claude-codetheir own version, first seen Sep 20, 2026
Other versions12 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 12
- Waishnav/devspace★ 5.4ksame name
- nimadorostkar/Claude-Skills-collection★ 33same name
- sebnado/worldbuild-bench★ 9same name
- haxqer/skills★ 4same name
- vlln/subagents-skill★ 4same name
- skrrt-sh/skills★ 3same name
- mteam88/pi-config★ 2same name
- ershiyidian/QuayMesh★ 1same name
- MaGe1993/Claude-Code-Base-Harness★ 1same name
- wtsi-hgi/agentskills★ 1same name
2 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 repository23 other files it gives its agents.
- AGENTS.mdAGENTS.md
- broSkill
- create-my-skillSkill
- glassmorphismSkill
- swiftui-designSkill
- humanizerSkill
- ponytail-auditSkill
- ponytail-debtSkill
- ponytail-gainSkill
- ponytail-helpSkill
- ponytail-reviewSkill
- ponytailSkill
- clean-up-git-historySkill
- commit-msgSkill
- diagnoseSkill
- fixSkill
- grill-meSkill
- implementSkill
- loopSkill
- update-docsSkill
- verify-appSkill
- sudo-ssh-authSkill
- web-app-secSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 12 edited versions we read keep these.
3
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 14
Never use the top tier models for subagents.
- Distinctive: line 23
Prefer an indefinite wait when supported.
- Distinctive: line 25
Do not alternate sleeps with status checks or use short polling loops.
“Common” is measured: kept by at least 6 in 10 of the 12 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
Subagents
Delegate main or independent subtasks to subagents.
- Pick the model by the tool you run in:
- OpenCode: Nemotron 3.5 Super VL Medium;
- Codex (ChatGPT): GPT-6 Luna High (general tasks) or GPT-6.1 Terra High (priority choice for computer use tasks);
- Cursor: Grok 4.7 Medium (regular speed).
- Never use the top tier models for subagents.
- Relaunch the subagent with maximum reasoning level available in case the subagent failed and you feel that it happened because of lack of reasoning.
- Give subagents narrow tasks and only the context they need.
- Nested subagents are allowed.
Waiting for subagents
- Continue independent work while subagents run. When only waiting remains, use completion notifications or a wait that returns as soon as a subagent finishes.
- Prefer an indefinite wait when supported. Otherwise, use the longest timeout allowed by the environment and active instructions, and wait again if it expires.
- Do not alternate sleeps with status checks or use short polling loops. Inspect status separately only when needed to make a decision or intervene.
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.