grill-with-docs · haowjy — Claude Code skill: what it does, how to install
Challenges project plans through sequential questioning, verifies terminology against project docs, and records decisions on the fly.
npx skills add haowjy/creative-writing-skills --skill grill-with-docs -a claude-codeWhat grill-with-docs does
The skill runs an intensive interview over the author's plan by stepping through a decision tree. It starts with high-leverage dependencies that block downstream choices. Instead of overwhelming you, it asks a small cluster of targeted questions per turn, suggests a recommended answer for each, and waits for your response.
Every term gets checked against existing project sources: kb/vocab.md, domain glossaries, conventions in AGENTS.md, and active notes in work/. When terminology clashes with documented concepts, the agent flags the conflict immediately and demands a canonical choice. When it needs to inspect broader project context, it reads target files directly or spawns a focused subagent to keep the main interview context uncluttered.
As choices stabilize, the skill records them straight into work/ and updates vocabulary files. This workflow demands an existing documentation layout with kb/ and work/ directories, otherwise the evidence lookups and automated notes have nowhere to land.
What it helps with
- stress-test a design or narrative plan before writing code
- catch terminology conflicts against
kb/vocab.mdandAGENTS.md - unravel decision dependencies one branch at a time
- record stabilized tradeoffs and rationale into
work/notes
Example prompts for the agent
- Here is my proposed feature plan, run grill-with-docs to challenge it.
- Question my proposal branch by branch and compare terms with kb/vocab.md.
How to install grill-with-docs
1Skills CLI (Vercel)
npx skills add haowjy/creative-writing-skills --skill grill-with-docs -a claude-code2Manually: put the files in place
~/.claude/skills/grill-with-docs/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/grill-with-docs/
1Skills CLI (Vercel)
npx skills add haowjy/creative-writing-skills --skill grill-with-docs -a codex2Manually: put the files in place
.agents/skills/grill-with-docs/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add haowjy/creative-writing-skills --skill grill-with-docs -a cursor2Manually: put the files in place
.cursor/skills/grill-with-docs/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install haowjy/creative-writing-skills grill-with-docs2Manually: put the files in place
.github/skills/grill-with-docs/SKILL.md1Skills CLI (Vercel)
npx skills add haowjy/creative-writing-skills --skill grill-with-docs -a gemini-cli2Platform CLI
gemini skills install https://github.com/haowjy/creative-writing-skills --path cw/skills/grill-with-docsManually: put the files in place
.devin/skills/grill-with-docs/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add haowjy/creative-writing-skills --skill grill-with-docs -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about grill-with-docs
How to install grill-with-docs in Claude Code?
Run npx skills add haowjy/creative-writing-skills --skill grill-with-docs -a claude-code.
What does grill-with-docs do?
It conducts a structured interrogation of your proposed plan, points out conflicting terms against your knowledge base, and writes down resolved decisions.
What folder structure does grill-with-docs expect?
It checks terminology and stores notes using kb/vocab.md, AGENTS.md, and the work/ directory.
Source text
SKILL.md · Apache-2.0
--- name: grill-with-docs description: Use when challenging a plan — grills the author against documented decisions and sharpens terminology. --- # Grill With Docs Load `/intent-modeling` if it isn't already loaded. Interview the author relentlessly about every aspect of their plan until you reach a shared understanding. Walk down each branch of the decision tree, resolving dependencies between decisions one by one. Keep each turn to the next unresolved branch or a small cluster of related questions the author can answer in one reply, then wait. Provide your recommended answer for each question. ## Starting the Session Read the plan or proposed direction the author provides. Identify: - **Decision branches** — choices that gate other choices - **Dependency order** — which decisions must resolve first - **Terms needing precision** — vague, overloaded, or undefined vocabulary - **Assumptions** — things treated as given that may not be Start with the highest-leverage unresolved dependency — the earliest branch point that unblocks the most downstream decisions. ## Challenging Terminology Check every significant term against the project's existing vocabulary where it exists: 1. Vocabulary pages in the kb — `kb/vocab.md` for project-wide terms, `kb/<domain>/vocab.md` for domain-specific ones. 2. Project conventions in `AGENTS.md` — established names and labels. 3. Active work notes in `work/` — terms already defined for the current effort. 4. Prior decisions — terms established by earlier choices. When the author's language conflicts with documented terms, call it out immediately. Name the conflict: "You said X, but the kb defines Y for this concept — which should we use?" Sharpen vague or overloaded terms into canonical terms. Use concrete scenarios and edge cases to force precise boundaries — "Does 'sync' here mean the timeline beat, the character's realization, or something new?" See `/shared-dao` for the vocabulary discipline. ## Gathering Evidence When answering requires cross-referencing project materials beyond what is already in context, read the specific files directly, or spawn a focused subagent to gather evidence so your own context stays on the grilling conversation. Scope each lookup to one question and target specific files or directories — chapters, kb entries, outlines, prior decisions. Frame the lookup as an evidence-seeking question with file/path targets. Act on findings to sharpen your next question or to challenge the author's assumptions with evidence. ## Updating Documentation Update artifacts inline as decisions crystallize. Reasoning flattens the longer you wait — capture it in the moment. ### Active work notes Record decisions in `work/` as they stabilize — the decision, the reasoning, and any constraints that emerged. Keep open questions visible. ### KB vocabulary When the session produces a new canonical term or refines an existing one, update the appropriate vocabulary page: `kb/vocab.md` for cross-cutting terms, `kb/<domain>/vocab.md` for domain-specific ones. Follow `/kb-management` for page structure. ### Durable decisions Create durable decision records sparingly — only when the choice is hard to reverse, surprising without context, and involves a real tradeoff. Record it in the relevant kb page. Keep provisional decisions in `work/` until they prove durable. ## Session Rhythm Each cycle through a decision branch: 1. State the branch and why it matters now (dependency context). 2. Ask focused questions with your recommended answer for each. 3. Wait for the author's response. 4. If the answer raises a sub-question, drill into it before moving on. 5. If the answer can be verified against project materials, check the relevant files and use what you find to confirm or challenge. 6. When the branch resolves, update the relevant documentation surface immediately. 7. Advance to the next branch. The session ends when all branches of the decision tree are resolved and the documentation reflects the shared understanding.
© haowjy · Apache-2.0 · haowjy/creative-writing-skills
Similar skills
Everything for Writing and docs →grill-with-docs by other authors comes first.