brain
/brain
Entry point for real work. Use when a task touches more than one file, changes a shared signature, needs a design choice, or has an unknown cause. Skip it for small tasks.
Assessment and security
- Files in this package
- 5
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 10.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Entry point for real work. Use when a task touches more than one file, changes a shared signature, needs a design choice, or has an unknown cause. Skip it for small tasks.
Original author description
Entry point for real work. Use when a task touches more than one file, changes a shared signature, needs a design choice, or has an unknown cause. Skip it for small tasks.
How it works
SKILL.md has no level-two headings · 13 lines, 186 words
SKILL.md preview
--- name: brain description: Entry point for real work. Use when a task touches more than one file, changes a shared signature, needs a design choice, or has an unknown cause. Skip it for small tasks. --- # brain pick the shape of the task, do it small, prove it, close it. 1. restate the goal in one line. read the project `CLAUDE.md` and the `handoff.md` it points to (the `docs:` line says where). 2. pick the shape, then read its playbook from this skill's `playbooks/` folder: `new-project.md` (an idea with no design yet, or an empty repo), `feature.md`, `bug.md` or `investigation.md`. follow its steps in order. a step you skip is said as skipped, with the reason. - **project override.** if `<docs>/playbooks/<shape>.md` exists (`docs:` from the project `CLAUDE.md`), read it too. each line is `after "<base step, quoted exactly>": <what to do>` or `before "<base step>": <what to do>`. do that at that point. a quote that matches no base step is a broken override: say so and carry on with the base steps. 3. a task with several independent pieces, or one that needs a fresh pair of eyes, is delegated. follow `kstack:principle-delegate`. 4. before saying done, follow `kstack:principle-prove-it`. then run `kstack:close`. the user's `CLAUDE.md` wins over anything here. no is an acceptable answer to a request that doesn't hold up.
Files in this package (5)
- SKILL.md 1 KB
- playbooks/bug.md 1 KB
- playbooks/feature.md 1 KB
- playbooks/investigation.md 1 KB
- playbooks/new-project.md 1 KB
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · gmdkaio/kstack | 86 | MIT |
| jeremylongshore/tons-of-skills-marketplace | 100 | MIT |
| coco-research/coco | 100 | MIT |
| jeremylongshore/bobs-big-brain-registrar | 100 | Apache-2.0 |
| onetest-ai/echo | 100 | AGPL-3.0 |
| johantor/brain | 100 | Apache-2.0 |
| guangshinhaha/shinstack | 100 | MIT |
| atlasatakahraman/knowledge-skeleton | 100 | GPL-3.0 |
| kayossouza/brain-template | 100 | MIT |
Database snapshot: 13.11.2026 · package version 2ba32bd36fc7 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100systematic-debugging
Use when an observed failure, regression, flake, unexpected output, or unknown root cause needs investigation; a known-cause execution-ready fix remains Ralph. - 100writing-plans
Use when implementation needs dependency ordering, shared interface decisions, or a durable handoff; also for explicit plan requests. Skip mechanical tasks with obvious ordering. - 100brain-write-script
Write or change a script in this brain's scripts/ when a query or API call is worth keeping instead of regenerating, or a system needs probing, or connect a system through an MCP server in .mcp.json. Places it by the system it touches, keeps it read-only unless its name says… - 9008-debug
Reproduce and fix a known bug, or find an unknown root cause by hypothesis validation. Use when the user wants to fix a bug, find why something breaks, or reopen a stuck investigation. Not for building a feature or reviewing a diff. - 90incremental-implementation
Implement multi-file changes in thin vertical slices—implement, test, verify, commit. Use when a feature touches more than one file or feels too large for one pass. Triggers: "incremental", "vertical slice", "one step at a time", "small commits". - 100nerd-diagnose
Use when establishing why behavior is broken, unexpected, or inconsistent and reporting a confirmed, probable, or unknown cause without performing repair.