grill-brainstorm-build
Use when the user wants to go from a rough idea to implementation, running through interrogation, options exploration, and build in one flow.
Skill by hjr15
Chalk Labs
Need help deploying grill-brainstorm-build?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/hjr15-grill-brainstorm-build
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design, productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/grill-brainstorm-build
What's inside
- Skill name
grill-brainstorm-build- Description
- Use when the user wants to go from a rough idea to implementation, running through interrogation, options exploration, and build in one flow. Triggers on "grill me then build", "design-to-build", or "think it through then implement".
- License
- MIT
- Instructions
- 30 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Grill → Brainstorm → Build ## Overview A three-phase pipeline that takes a rough design idea through structured interrogation, options exploration, and implementation — without requiring the user to manually chain skill invocations. ## When to Use - User says "grill me on this then build it" - User says "think it through then implement" - User says "design-to-build" or "full flow" - User is starting a non-trivial feature and has a rough idea but no firm spec - **Alignment-only mode:** user says "grill me so we're aligned", "firm up my understanding", or "find the drift" — the goal is comprehension, not a build (see Notes) ## Steps ### Phase 1: Grill Invoke the `superpowers:brainstorming` skill (or run an inline interrogation if the plugin is unavailable: ask the user 5+ probing questions about scope, constraints, edge cases, success criteria, and out-of-scope items before proceeding to options exploration). Do not proceed to Phase 2 until the user has answered all branches and confirmed they are satisfied with the interrogation. ### Phase 2: Brainstorm Invoke the `superpowers:brainstorming` skill using the outputs of Phase 1 as context. Surface at least 3 distinct implementation options. Do not proceed until the user selects or synthesizes a direction. ### Phase 3: Build Invoke `superpowers:writing-plans` to produce a structured plan, then hand off to `superpowers:subagent-driven-development` for execution. Apply `superpowers:verification-before-completion` before declaring done. ## Notes - Never collapse phases. Each phase requires explicit user confirmation before proceeding. - If user skips Phase 1 explicitly, note the skipped interrogation and proceed. - **Product before schedule.** Lead the grill with product and technical-scope questions; defer schedule/timeline/weekly-hours/sequencing questions until the user raises them — asking them up front reads as off-focus even when a kickoff brief mandated them. - **De-jargon multiple-choice options.** When a grill question offers options phrased in technical or business jargon, put a one-line plain-language "why this matters" up front, not buried in the option description. - **Alignment-only mode** — when the goal is to firm up understanding / find drift, not build: state up front that you're using the grill methodology and dropping Phases 2–3. **Ground-truth first** — dispatch Explore agents / read the real current code + docs so you grill against reality, not assumptions (drift is the whole point). Answer the user's "where is this documented?" sub-questions with file paths. Stop at Phase 1; only continue to Brainstorm/Build if the user explicitly asks.
From hjr15/claude-kit, MIT.
More from hjr15/claude-kit
- 0.524score
architect-team
hjr15
Deploy the architecture agents (architect, code-architect, backend-architect) as a coordinated design council to design or review a feature/system before it's…
SkillBFreshness: BClaude Code +9 - 0.524score
code-team
hjr15
Deploy the code-quality agents (code-quality-reviewer, code-architect) alongside the /code-review skill to answer "is this code well-built?" — clean code…
SkillBFreshness: BClaude Code +9 - 0.524score
doc-accuracy-audit
hjr15
Use when validating that a repo's READMEs and docs are still accurate against the current code — endpoints, ports, counts, env vars, commands, schemas, event…
SkillBFreshness: BClaude Code +9 - 0.524score
backport-review-fix-into-plan-of-record
hjr15
Use when a per-task review during subagent-driven development catches and fixes a real bug in code that a plan-of-record authored — edit the plan itself to…
SkillBFreshness: BClaude Code +9 - 0.524score
fix-blocker-rather-than-override
hjr15
Use when a pre-existing master CI gate is red and blocking your unrelated PR — decide between fix-the-blocker (compounds across all PRs) and admin-override…
SkillBFreshness: BClaude Code +9 - 0.524score
generator-oracle-zero-diff-verify
hjr15
Use when building a code-generator, sanitizer, or export pipeline whose output already exists as a hand-maintained artifact — make zero-git-diff against that…
SkillBFreshness: BClaude Code +9 - 0.524score
multi-agent-branch-collision-recovery
hjr15
Use when a commit landed on the wrong branch because a parallel agent shifted the active branch in a shared working directory.
SkillBFreshness: BClaude Code +9 - 0.519score
ghcr-first-push-403-seed-and-link
hjr15
Use when a CI build succeeds but the GHCR push fails 403 Forbidden for a service whose package never existed, on a personal (non-org) account.
SkillBFreshness: BClaude Code +9 - 0.517score
symlink-vs-realfile-classify-before-deleting-fanout-dups
hjr15
Use before deleting "duplicates" during a skill/config dedup on a symlink-fanned tree — run a `test -L` classification loop first, so you remove only genuine…
SkillBFreshness: BClaude Code +9 - 0.517score
git-checkout-B-from-worktree-moves-a-foreign-checkouts-branch
hjr15
Use before `git checkout -B <branch>` / `git branch -f` / `git reset --hard` on a SHARED branch name from inside a worktree — `-B` force-moves the ref even…
SkillBFreshness: BClaude Code +9 - 0.474score
debugging-feedback-loop
hjr15
Use when debugging a bug you can't yet reproduce reliably, or when you need a tight/automated reproduction loop before hypothesising — especially…
SkillCFreshness: CClaude Code +9 - 0.474score
readme-diagram-audit
hjr15
Use when auditing a repo for diagram coverage and convention compliance — scans docs/diagrams/ for completeness, validates frontmatter against the taxonomy…
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Aug 15, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/hjr15/claude-kit
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from hjr15
- 0.524score
fix-blocker-rather-than-override
hjr15
Use when a pre-existing master CI gate is red and blocking your unrelated PR — decide between fix-the-blocker (compounds across all PRs) and admin-override…
SkillBFreshness: BClaude Code +9 - 0.524score
backport-review-fix-into-plan-of-record
hjr15
Use when a per-task review during subagent-driven development catches and fixes a real bug in code that a plan-of-record authored — edit the plan itself to…
SkillBFreshness: BClaude Code +9 - 0.524score
code-team
hjr15
Deploy the code-quality agents (code-quality-reviewer, code-architect) alongside the /code-review skill to answer "is this code well-built?" — clean code…
SkillBFreshness: BClaude Code +9 - 0.524score
architect-team
hjr15
Deploy the architecture agents (architect, code-architect, backend-architect) as a coordinated design council to design or review a feature/system before it's…
SkillBFreshness: BClaude Code +9 - 0.524score
doc-accuracy-audit
hjr15
Use when validating that a repo's READMEs and docs are still accurate against the current code — endpoints, ports, counts, env vars, commands, schemas, event…
SkillBFreshness: BClaude Code +9 - 0.524score
generator-oracle-zero-diff-verify
hjr15
Use when building a code-generator, sanitizer, or export pipeline whose output already exists as a hand-maintained artifact — make zero-git-diff against that…
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get grill-brainstorm-build running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
grill-brainstorm-build skill. Agent Mart. https://agentmart.dev/skills/hjr15-grill-brainstorm-build (last updated 2026-10-24)