task-brief
Skill1 starsChanged 2 months agoMITView on GitHub
Turn an implementation, debugging, audit, or migration request into a precise scope, objective, constraint, file, validation, and reporting brief before work begins. Use for multi-file, ambiguous, architectural, or staged work and explicit requests for a task brief. Do not use for architecture inventory, refactor execution, commit review, session handoffs, or trivial requests with complete acceptance criteria.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- arumaekawa/codex-skillstheir own version, first seen Oct 25, 2026
- THROVIA/Task_Brieftheir own version, first seen Oct 17, 2026
- Baba01hacker666/dummyskillstheir own version, first seen Oct 13, 2026
- zarubinvibe/themistheir own version, first seen Sep 8, 2026
- zarubinvibe/themiztheir own version, first seen Sep 8, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- zarubinvibe/themis★ 11same name
- zarubinvibe/themiz★ 11same name
- arumaekawa/codex-skills★ 4same name
- THROVIA/Task_Brief★ 4same name
- Baba01hacker666/dummyskills★ 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 repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- architecture-auditSkill
- commit-reviewSkill
- cxpSkill
- docs-updateSkill
- safe-refactorSkill
- session-handoffSkill
- cxpSkill
The file
We split this skill into the rules it gives an agent
8 rules. 0 are common, 8 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
8
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 15
Use when a request needs scope clarification, acceptance criteria, protected areas, phased work, or a validation contract before …
- Distinctive: line 20
Do not use for a full architecture inventory, behavior-preserving extraction, documentation editing, commit preparation, or handoff …
- Distinctive: line 44
Stop with the brief; do not implement it.
- Distinctive: line 56
Do not infer permission for product edits, provider calls, dependencies, Git mutations, or broader cleanup.
- Distinctive: line 58
Do not turn an uncertain dependency assumption into an architecture finding.
- Distinctive: line 59
Do not include secrets or credential values.
- Distinctive: line 63
Verify named existing paths before presenting them as current.
- Distinctive: line 65
Check that acceptance criteria cover the objective and that non-goals prevent scope drift.
“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
Task Brief
Purpose
Produce an implementation-ready contract without solving or implementing the task. Separate verified repository facts from assumptions and decisions.
Trigger
Use when a request needs scope clarification, acceptance criteria, protected areas, phased work, or a validation contract before implementation.
Do not trigger
Do not use for a full architecture inventory, behavior-preserving extraction, documentation editing, commit preparation, or handoff generation. Route those responsibilities to their dedicated skills.
Inputs
- User objective and acceptance criteria.
- Current branch, working-tree state, and relevant repository instructions.
- Known implementation evidence and unresolved assumptions.
- Allowed and protected product areas, files, dependencies, and provider use.
- Required final report or artifact format.
Workflow
- Inspect only enough repository context to define the task accurately.
- Record observed facts separately from assumptions and open decisions.
- State one concrete objective and measurable acceptance criteria.
- Mark affected areas and explicitly unchanged areas: frontend, backend, prompts, graph, RAG, memory, persistence, providers, tests, dependencies, configuration, and deployment.
- List expected files, protected files, constraints, and non-goals.
- Identify validation appropriate to the proposed scope using
AGENTS.md. - Define the required final report, approval boundary, risks, and rollback point.
- Stop with the brief; do not implement it.
Outputs
Return a concise brief using the task brief format. Call out any missing decision that genuinely blocks safe implementation.
Safety
- Remain read-only unless the user separately authorizes writing the brief to a named path.
- Do not infer permission for product edits, provider calls, dependencies, Git mutations, or broader cleanup.
- Do not turn an uncertain dependency assumption into an architecture finding.
- Do not include secrets or credential values.
Validation
- Verify named existing paths before presenting them as current.
- Label planned paths and commands as planned.
- Check that acceptance criteria cover the objective and that non-goals prevent scope drift.
- Confirm that the validation plan and final report match the requested scope.
References
AGENTS.md— durable repository rules and validation.docs/agents/README.md— tooling ownership.- Task brief format — output structure.
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.

