grill-me
Skill4 starsChanged 3 months agoApache-2.0View on GitHub
Interview the user relentlessly — one question at a time, with a recommended answer for each — to stress-test a plan, design, process, or decision until every branch is resolved. Checkpoints each answer to a dated file in brainstorms/, so the session survives context drift and can resume where it left off after an interruption. Use when the user wants to think through or pressure-test something, get grilled on a design or decision, or says "grill me".
193 other projects use a version of this, 67 of them edited.
Adopted by
Most recently taken up by
- xiaoxiaosrm/dsh-mattpocock-skillstheir own version, first seen Nov 8, 2026
- ishandutta2007/Awesome-Claude-Skillstheir own version, first seen Nov 8, 2026
- microsoft/frontier-agentic-modernization-rvastheir own version, first seen Nov 8, 2026
- The-Vibe-Company/quivrtheir own version, first seen Nov 8, 2026
- anotherben/claude-harnessedited, first seen Nov 8, 2026
- codedb-io/codedb-lenstheir own version, first seen Nov 8, 2026
Used with
Projects with it also use brief-to-tasks, writing-beats and 9 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
julianoczkowskiSkillbrief-to-tasks
Break a design brief into an ordered checklist of independently buildable tasks using vertical slices. Saves as a markdown checklist. Use when user wants to break down work, create tasks from a brief, plan implementation order, or mentions "tasks" or "breakdown".
Also in 7 projects that use this
LesaniSkillwriting-beats
Shape an article as a journey of beats, choose-your-own-adventure style. The user picks a starting beat from the raw material, you write only that beat, then offer options for where to pivot next, beat by beat, until the article reaches a natural end. Use when the user has raw material and wants to assemble it as a narrative rather than an argument.
Also in 66 projects that use this
simo97Skillrequest-refactor-plan
Use this skill when user wants to create a refactor plan.
Also in 43 projects that use this
simo97Skillmigrate-to-shoehorn
Migrate test files from `as` type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace `as` in tests, or needs partial test data.
Also in 71 projects that use this
LesaniSkillwriting-shape
Take a markdown file of raw material and shape it into an article through a conversational session — drafting candidate openings, growing the piece paragraph by paragraph, arguing about format (lists, tables, callouts, quotes) at each step. Use when the user has a pile of notes, fragments, or a rough draft and wants help turning it into something publishable.
Also in 65 projects that use this
liufeng1976Skillsetup-ts-deep-modules
Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.
Also in 42 projects that use this
- +2 more from simo97
Other versions193 projects use a version of this.
Word for word · 0
None found.
Edited · 67
- langfuse/langfuse★ 37k24% same
- humanlayer/humanlayer★ 13k22% same
- openstatusHQ/openstatus★ 9.3k27% same
- webiny/webiny-js★ 8k25% same
- sanity-io/sanity★ 6.4k27% same
- camunda/camunda★ 4.4k27% same
- event-catalog/eventcatalog★ 3k25% same
- Akylas/OSS-DocumentScanner★ 2.6k25% same
- openstatusHQ/data-table-filters★ 2.3k27% same
- owid/owid-grapher★ 1.6k26% same
57 more not listed
Wrote their own version · 126
- mattpocock/skills★ 309ksame name
- QuivrHQ/quivr★ 40ksame name
- The-Vibe-Company/quivr★ 40ksame name
- nrwl/nx★ 29ksame name
- windmill-labs/windmill★ 18ksame name
- GreyDGL/PentestGPT★ 16ksame name
- jaredpalmer/kev★ 9.8ksame name
- ArcReel/ArcReel★ 6.3ksame name
- vinvcn/mattpocock-skills-zh-CN★ 5.5ksame name
- pollinations/pollinations★ 5.3ksame name
116 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 repository25 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- adr-writingSkill
- authoring-goal-contractsSkill
- collaborating-asyncSkill
- committing-changesSkill
- cross-agent-dispatchSkill
- documenting-decisionsSkill
- feature-status-registrySkill
- grill-yourselfSkill
- handling-micro-fixesSkill
- multi-instance-dispatchSkill
- notifying-the-developerSkill
- orchestrating-goal-loopsSkill
- orchestrating-lean-dispatchSkill
- performance-playbookSkill
- python-project-patternsSkill
- recovering-from-failuresSkill
- running-build-checkpointsSkill
- searching-prior-artSkill
- security-checklistSkill
- selecting-review-gatesSkill
- severity-calibrationSkill
- syncing-framework-docsSkill
- testing-playbookSkill
- wrapping-up-sessionsSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 177 edited versions we read keep these.
5
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 12
Ask exactly one question at a time, and wait for my answer before asking the next.
- Distinctive: line 14
If a question can be answered by reading the codebase or existing project files, do that instead of asking me.
- Distinctive: line 15
If you hit a branch I can't or shouldn't answer — it's someone else's call, or needs data I don't have — don't guess.
- Distinctive: line 24
After each append, update Key decisions and Open threads to reflect the new state.
- Distinctive: line 59
If no match exists: create a new file from the template above and start fresh.
“Common” is measured: kept by at least 6 in 10 of the 177 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
Purpose: turn a fuzzy idea into resolved, captured decisions by interviewing the user relentlessly and checkpointing every answer to disk — so the session survives context drift and can resume after an interruption.
The interview
Interview me relentlessly about every aspect of this topic until we reach a shared understanding. Walk down each branch of the decision tree, resolving dependencies between decisions one at a time.
- Ask exactly one question at a time, and wait for my answer before asking the next.
- For every question, propose your recommended answer with a one-line rationale, so I can react ("yes", or "yes, but change X") instead of authoring from a cold start.
- If a question can be answered by reading the codebase or existing project files, do that instead of asking me.
- If you hit a branch I can't or shouldn't answer — it's someone else's call, or needs data I don't have — don't guess. Record it under Flagged for others and move on.
- Match depth to stakes: a tight, well-understood topic may resolve in ~15 questions; a fuzzy or high-stakes one may take 50 or more. Don't wrap early — the edge cases that matter usually surface in the back half.
Checkpoint every answer to disk
The live conversation will drift as it grows; the file is the source of truth.
- The moment I answer a question, append that exchange to the session file (see below) before asking the next question.
- Capture my answer verbatim in the Q&A log — do not paraphrase it.
- After each append, update Key decisions and Open threads to reflect the new state.
Session file
Store one file per topic at brainstorms/YYYY-MM-DD-<topic-slug>.md, where the date is when the topic was first grilled and <topic-slug> is a short kebab-case name. Create the brainstorms/ folder and the file lazily, on the first answered question. Use this structure:
# Grill: <Topic>
> Status: in-progress | complete
> Started: YYYY-MM-DD · Last updated: YYYY-MM-DD
## Highlights
<3–6 lines a future reader can skim in 30 seconds>
## Key decisions
- <Decision stated as a commitment — what we settled and why>
## Open threads
- <An unresolved branch, or the exact next question to ask on resume>
## Flagged for others
- <A question I couldn't or shouldn't answer — and who/what it needs>
## Q&A log
### Session 1 — YYYY-MM-DD
**Q:** <your question>
**A:** <my verbatim answer>Resume an interrupted session
On startup, derive the topic slug from my request and search brainstorms/ for a file whose name ends with -<topic-slug>.md.
- If a match exists: read it. Briefly replay Highlights and Key decisions so we both re-anchor, tell me what's still in Open threads, and ask whether anything has changed since the last session. Then continue from the first open thread. Do not re-ask questions already answered in the Q&A log. Append new exchanges under a fresh
### Session N — YYYY-MM-DDmarker in the same file. - If no match exists: create a new file from the template above and start fresh.
When the session ends
Set Status: complete. Make sure Highlights works as a 30-second digest, Key decisions read as commitments, and Open threads / Flagged for others clearly state what still needs another human or more thought. Then offer to fold the captured decisions into the relevant skill, document, or plan.
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.

