just
Skill1 starsChanged 7 months agoMITView on GitHub
Use only when the user explicitly invokes just-mode or asks to skip ceremony, avoid research, avoid planning, avoid extra skills, avoid subagents, avoid explore agents, avoid TODOs, stop overthinking, or directly execute a simple requested action. Do not use by default or infer it from ordinary tasks.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- Recluse/HyperMnesiatheir own version, first seen Nov 5, 2026
- disler/bowsertheir own version, first seen Oct 8, 2026
- casey/justtheir own version, first seen Oct 8, 2026
- wyattjoh/skillstheir own version, first seen Sep 27, 2026
- asw101/ai-skillstheir own version, first seen Aug 21, 2026
- moabualruz/fulcrumtheir own version, first seen Apr 28, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- casey/just★ 37ksame name
- disler/bowser★ 265same name
- Recluse/HyperMnesia★ 17same name
- moabualruz/fulcrum★ 4same name
- asw101/ai-skills★ 3same name
- wyattjoh/skills★ 3same name
- CodeWithBehnam/cc-docs★ 2same 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.
The file
We split this skill into the rules it gives an agent
11 rules. 0 are common, 11 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
11
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 16
Use this skill only when the user explicitly says something like:
- Distinctive: line 23
Do not activate this skill merely because the task looks simple.
- Distinctive: line 29
Do not brainstorm, write specs, enter plan mode, or create TODO lists.
- Distinctive: line 30
Do not dispatch subagents, Explore agents, research agents, review agents, or parallel agents.
- Distinctive: line 31
Do not perform broad codebase exploration.
- Distinctive: line 32
Do not ask clarifying questions unless the request is impossible or dangerous without one.
- Distinctive: line 33
Do not invoke other process skills unless required by a higher-priority system/developer rule.
- Distinctive: line 35
Verify with the fastest command that proves the change worked.
- Distinctive: line 49
If verification is impossible or would be too expensive, say exactly that and give the next best evidence.
- Distinctive: line 53
Use terse status updates only when needed.
- Distinctive: line 60
Avoid long explanations, alternatives, design discussion, or summaries of obvious steps.
“Common” is measured: kept by at least 6 in 10 of the 5 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
Just
Overview
Just mode is an anti-ceremony workflow for agents. When the user explicitly asks for it, do the requested work directly with the smallest safe action path.
Core principle: obey the user's request for low ceremony without becoming reckless.
Activation
Use this skill only when the user explicitly says something like:
/just,just,just mode直接做,直接改,开干,别废话不要调研,不要 planning,不要 spec,不要 TODO别调用其他 skill,别走流程,stop overthinking
Do not activate this skill merely because the task looks simple. If the user did not ask for just-mode behavior, use the normal workflow.
Rules
When active:
- Do not brainstorm, write specs, enter plan mode, or create TODO lists.
- Do not dispatch subagents, Explore agents, research agents, review agents, or parallel agents.
- Do not perform broad codebase exploration; inspect only the exact files, commands, or state needed for the task.
- Do not ask clarifying questions unless the request is impossible or dangerous without one.
- Do not invoke other process skills unless required by a higher-priority system/developer rule.
- Make the smallest change that satisfies the request.
- Verify with the fastest command that proves the change worked.
- Report only what changed and the verification result.
Safety Boundary
Just mode does not authorize destructive or externally visible actions.
Still ask before:
- deleting data, files, branches, databases, or infrastructure
- force pushing, resetting, cleaning, or overwriting uncommitted work
- publishing, sending messages, opening/closing PRs or issues
- changing secrets, permissions, billing, or shared production systems
If verification is impossible or would be too expensive, say exactly that and give the next best evidence.
Response Style
Use terse status updates only when needed. Final response should be short:
已完成：<what changed>。
验证：<command/result>。Avoid long explanations, alternatives, design discussion, or summaries of obvious steps.
Common Mistakes
| Mistake | Correct behavior |
| Treating just mode as permission to skip safety | Keep safety boundaries |
| Running broad research first | Inspect only what is necessary |
| Creating a plan or TODO list | Directly edit or execute |
| Asking preference questions | Pick the obvious minimal path |
| Invoking every relevant skill | Suppress process skills unless higher-priority rules force them |
| Dispatching subagents or Explore agents | Work in the current agent with targeted reads/searches only |
| Starting broad codebase discovery | Use the narrowest file read or search needed |
| Reporting a long narrative | Give changed/result only |
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.