delegate-task
Skill4 starsChanged 6 months agoMITView on GitHub
Delegate tasks to OpenSpace — a full-stack autonomous worker for coding, DevOps, web research, and desktop automation, backed by an extensive MCP tool and skill library. Skills auto-improve through use, reducing token consumption over time. A cloud community lets agents share and collectively evolve reusable skills.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- danielvm-git/bigpowerstheir own version, first seen Nov 6, 2026
- bradautomates/second-braintheir own version, first seen Oct 9, 2026
- Samin12/skillsword for word, first seen Aug 14, 2026
- leavesprior/claude-skillstheir own version, first seen Feb 22, 2026
Other versions4 projects use a version of this.
Word for word · 1
- Samin12/skills★ 8100% same
Edited · 0
None found.
Wrote their own version · 3
- danielvm-git/bigpowers★ 288same name
- bradautomates/second-brain★ 217same name
- leavesprior/claude-skills★ 0same name
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 repository79 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- 21st-magicSkill
- alpacaSkill
- ask-mattSkill
- autoresearchSkill
- auto-stage-commitSkill
- browser-harnessSkill
- codebase-designSkill
- codeburnSkill
- codex-reviewSkill
- compact-memorySkill
- connect-recommendSkill
- connect-required-verification-informationSkill
- decision-mappingSkill
- design-an-interfaceSkill
- design-mdSkill
- diagnosing-bugsSkill
- domain-modelingSkill
- edit-articleSkill
- extract-designSkill
- ffmpegSkill
- find-skillsSkill
- firecrawlSkill
- generateSkill
- git-guardrails-claude-codeSkill
- grillingSkill
- grill-me-codexSkill
- grill-meSkill
- grill-with-docs-codexSkill
- grill-with-docsSkill
- gwSkill
- handoffSkill
- implementSkill
- improve-codebase-architectureSkill
- kie-aiSkill
- lightragSkill
- llmfitSkill
- loop-meSkill
- migrate-to-shoehornSkill
- notebooklmSkill
- obsidian-vaultSkill
- openspaceSkill
- playwrightSkill
- prototypeSkill
- qaSkill
- request-refactor-planSkill
- resolving-merge-conflictsSkill
- reviewSkill
- roastSkill
- scaffold-exercisesSkill
- session-handoffSkill
- setup-matt-pocock-skillsSkill
- setup-pre-commitSkill
- site-teardownSkill
- skill-discoverySkill
- skilluiSkill
- spline-3dSkill
- stripe-appsSkill
- stripe-best-practicesSkill
- stripe-directorySkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 33
Check response for evolved_skills.
- Distinctive: line 53
Search for available skills before deciding whether to handle a task yourself or delegate.
- Very specific: line 130
If execute_task times out, first check the host's MCP timeout settings.
- Distinctive: line 132
After every OpenSpace call, tell the user what happened: task result, any evolved skills, and your upload decision.
“Common” is measured: kept by at least 6 in 10 of the 3 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
Delegate Tasks to OpenSpace
OpenSpace is connected as an MCP server. Whether the host uses stdio, sse, or streamable-http, you have the same 4 tools available: execute_task, search_skills, fix_skill, upload_skill.
When to use
- You lack the capability — the task requires tools or capabilities beyond what you can access
- You tried and failed — you produced incorrect results; OpenSpace may have a tested skill for it
- Complex multi-step task — the task involves many steps, tools, or environments that benefit from OpenSpace's skill library and orchestration
- User explicitly asks — user requests delegation to OpenSpace
Tools
execute_task
Delegate a task to OpenSpace. It will search for relevant skills, execute, and auto-evolve skills if needed.
execute_task(task="Monitor Docker containers, find the highest memory one, restart it gracefully", search_scope="all")| Parameter | Required | Default | Description |
task | yes | — | Task instruction in natural language |
search_scope | no | "all" | Local + cloud; falls back to local-only if no API key |
max_iterations | no | 20 | Max agent iterations — increase for complex tasks, decrease for simple ones |
Check response for evolved_skills. If present with upload_ready: true, decide whether to upload (see "When to upload" below).
{
"status": "success",
"response": "Task completed successfully",
"evolved_skills": [
{
"skill_dir": "/path/to/skills/new-skill",
"name": "new-skill",
"origin": "captured",
"change_summary": "Captured reusable workflow pattern",
"upload_ready": true
}
]
}search_skills
Search for available skills before deciding whether to handle a task yourself or delegate.
search_skills(query="docker container monitoring", source="all")| Parameter | Required | Default | Description |
query | yes | — | Search query (natural language or keywords) |
source | no | "all" | Local + cloud; falls back to local-only if no API key |
limit | no | 20 | Max results |
auto_import | no | true | Auto-download top cloud skills locally |
fix_skill
Manually fix a broken skill.
fix_skill(
skill_dir="/path/to/skills/weather-api",
direction="The API endpoint changed from v1 to v2, update all URLs and add the new 'units' parameter"
)| Parameter | Required | Description |
skill_dir | yes | Path to skill directory (must contain SKILL.md) |
direction | yes | What's broken and how to fix — be specific |
Response has upload_ready: true → decide whether to upload.
upload_skill
Upload a skill to the cloud community. For evolved/fixed skills, metadata is pre-saved — just provide skill_dir and visibility.
upload_skill(
skill_dir="/path/to/skills/weather-api",
visibility="public"
)For new skills (no auto metadata — defaults apply, but richer metadata improves discoverability):
upload_skill(
skill_dir="/path/to/skills/my-new-skill",
visibility="public",
origin="imported",
tags=["weather", "api"],
created_by="my-bot",
change_summary="Initial upload of weather API skill"
)| Parameter | Required | Default | Description |
skill_dir | yes | — | Path to skill directory (must contain SKILL.md) |
visibility | no | "public" | "public" or "private" |
origin | no | auto | How the skill was created |
parent_skill_ids | no | auto | Parent skill IDs |
tags | no | auto | Tags |
created_by | no | auto | Creator |
change_summary | no | auto | What changed |
When to upload
| Situation | Action |
| Skill was originally from the cloud | Upload back as "public" — return the improvement to the community |
| Fix/evolution is generally useful | Upload as "public" |
| Fix/evolution is project-specific | Upload as "private", or skip |
| User says to share | Upload with the visibility the user wants |
Notes
execute_taskmay take minutes — this is expected for multi-step tasks.- If
execute_tasktimes out, first check the host's MCP timeout settings. Changing fromstdioto HTTP (sseorstreamable-http) does not remove host-side per-call time limits. upload_skillrequires a cloud API key; if it fails, the evolved skill is still saved locally.- After every OpenSpace call, tell the user what happened: task result, any evolved skills, and your upload decision.
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.

