Chalk Labs
Need help deploying lv-brainstorm?
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/slicedbread-marc-lv-brainstorm
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
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/lv-brainstorm
What's inside
- Skill name
lv-brainstorm- Description
- Capture and explore ideas. Creates or updates briefs in plans/briefs/. Use when the user wants to brainstorm, explore an idea, or add to the backlog.
- License
- MIT
- Instructions
- 75 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Brainstorm Role **`help` / `--help`:** if that is the whole argument, print the `## Usage` block at the end of this file verbatim and stop. You are in **brainstorm mode**. Your job is to capture ideas, explore options, and help the user think through problems — NOT to write implementation steps or edit source code. ## What you do 1. **Assign a brief ID** — when the user describes something they want to build or change: ```bash new_id=$(scripts/lv-exec.sh lv-counter-next.sh BRF) ``` This reads the counter, prints the prefixed ID (e.g., `BRF-021`), and increments. Create a brief in `plans/briefs/$new_id-<name>.md` following `plans/briefs/TEMPLATE.md`. New briefs must include `schema_version: 4` in their header. **Add its INDEX.md row in the same edit — never "later".** A brief file with no row in `plans/briefs/INDEX.md` is invisible: it is not in the backlog you display on startup, not in the board's brainstorm count, and not in anything `/lv-spec` reads. client-e accumulated 19 of them. The row is `- [<file>.md](<file>.md) — <one-line description>` under the section matching the brief's status. 2. **Capture ideas** — fill in Problem, Rationale, and initial approach from the conversation 3. **Explore options** — help the user think through tradeoffs, list alternatives, surface risks 4. **Track status** — set the brief status appropriately: - `Idea` — just captured, no exploration yet - `Exploring` — actively discussing options - `Decided` — user has confirmed an approach, ready for planning - `Parked` — nobody is carrying it forward. Never set this by hand: `scripts/lv-exec.sh lv-brief-park.sh BRF-NNN "<why>"` stamps the dated reason and moves the INDEX.md row, and `--clear` reverses it. Offer it when the user says an idea is dead, superseded, or "not now" — a parked brief stops counting as an incomplete brainstorm while staying on the record. 5. **Maintain the index** — after creating or updating any brief, update `plans/briefs/INDEX.md` to reflect the current state. Move items between sections as their status changes. `Planned` is a waypoint, not a resting place: once a brief's plan reaches `complete` in `plans/REGISTRY.md` (check with `grep`), **remove that brief's entry from INDEX.md** rather than leaving it under `Planned` — a brief with a shipped plan is not backlog, and `Planned` filling up with terminal entries buries anything still live under it (client-c WFI-028). The brief file itself and `plans/REGISTRY.md` keep the historical record; INDEX.md only needs to reflect what is still ahead. 6. **Commit** — when done exploring or deciding, commit changes: ``` scripts/lv-exec.sh lv-commit.sh "brainstorm: BRF-NNN-<brief-name> — <Idea|Exploring|Decided>" --add plans/briefs/ ``` ## Rules - **Do NOT** edit source code files (src/,tests/) - **Do NOT** write implementation steps — that's the planner's job - **Do NOT** make decisions for the user — present options and let them choose - Briefs are living documents — rewrite freely until status is `Decided` - When a brief reaches `Decided`, it becomes input for a `/lv-spec` session - When a brief reaches `Planned`, add the plan link and move it in the index - When that plan reaches `complete`, remove the brief's entry from the index — `Planned` holds work waiting to ship, not a shipped history - Every brief has a row in INDEX.md from the moment it exists — creating one without a row is the one mistake in this skill nothing downstream can recover from - An idea nobody will pick up gets **parked**, not left at `Exploring` forever — `lv-brief-park.sh` — and never deleted - Quick ideas can start as lightweight entries (just Problem + a few sentences); they don't need every template section filled in immediately ## On startup Run these three commands in parallel before reading further: ```bash scripts/lv-exec.sh lv-branch-check.sh develop ``` ```bash cat plans/briefs/INDEX.md ``` ```bash scripts/lv-exec.sh lv-list-brainstorms.sh --drift ``` If branch check exits 1, stop and tell the user: "This skill must run on `develop`. Run `git checkout develop` first." Display the current backlog from INDEX.md as a summary table. Then tell the user: "Run `/model sonnet`, then describe what you'd like to explore or create." If `--drift` listed anything, those brief files exist with no row in INDEX.md — invisible to the backlog above and to every other surface. Say how many, and offer the repair once: `scripts/lv-exec.sh lv-brainstorm-reindex.sh` (dry run) then `--apply`. Don't run it unasked — it writes to the user's backlog. ## Committing work After creating or updating a brief, commit: ``` scripts/lv-exec.sh lv-commit.sh "brainstorm: <brief-name> — <Idea|Exploring|Decided>" --add plans/briefs/ ``` ## Usage ```text /lv-brainstorm show the backlog, then capture or explore an idea /lv-brainstorm <idea> start straight on that idea (a new or existing brief) ```
From slicedBread-marc/levain, MIT.
More from slicedBread-marc/levain
- 0.524score
lv-bug
slicedBread-marc
File a new bug report.
SkillBFreshness: BClaude Code +9 - 0.524score
lv-dep-notify
slicedBread-marc
File a notice, request or bug into another project's own pipeline.
SkillBFreshness: BClaude Code +9 - 0.524score
lv-rollback
slicedBread-marc
Execute a rollback for a completed or active plan.
SkillBFreshness: BClaude Code +9 - 0.524score
lv-consistency
slicedBread-marc
Cross-plan consistency pass.
SkillBFreshness: BClaude Code +9 - 0.521score
lv-stage
slicedBread-marc
Start or stop the release-branch staging environment.
SkillBFreshness: BClaude Code +9 - 0.521score
lv-help
slicedBread-marc
List every skill with its one-line purpose, or one skill's usage.
SkillBFreshness: BClaude Code +9 - 0.521score
lv-orchestrate
slicedBread-marc
Drive one bug, brief, or plan through the pipeline unattended, or start/stop the dispatcher daemon.
SkillBFreshness: BClaude Code +9 - 0.521score
lv-claim
slicedBread-marc
Take, release, or inspect the claim on a plan, bug, or brief — the lease that keeps another session off an artifact you are driving.
SkillBFreshness: BClaude Code +9 - 0.521score
lv-rack
slicedBread-marc
The cross-project board — every coordinated project's pipeline in one grid, and what waits on a human.
SkillBFreshness: BClaude Code +9 - 0.516score
lv-chrome-test
slicedBread-marc
Alias: now /lv-test <ID> browser.
SkillBFreshness: BClaude Code +9 - 0.516score
lv-debug
slicedBread-marc
Alias: now /lv-test (manual walkthrough) and /lv-bug.
SkillBFreshness: BClaude Code +9 - 0.515score
lv-attend
slicedBread-marc
Drain the orchestrator's gate queue — approve specs, run manual acceptance criteria, unstick parked plans.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 24, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/slicedBread-marc/levain
- Website
- slicedbread.ca/levain
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from slicedBread-marc
- 0.524score
lv-rollback
slicedBread-marc
Execute a rollback for a completed or active plan.
SkillBFreshness: BClaude Code +9 - 0.524score
lv-bug
slicedBread-marc
File a new bug report.
SkillBFreshness: BClaude Code +9 - 0.524score
lv-dep-notify
slicedBread-marc
File a notice, request or bug into another project's own pipeline.
SkillBFreshness: BClaude Code +9 - 0.524score
lv-consistency
slicedBread-marc
Cross-plan consistency pass.
SkillBFreshness: BClaude Code +9 - 0.521score
lv-help
slicedBread-marc
List every skill with its one-line purpose, or one skill's usage.
SkillBFreshness: BClaude Code +9 - 0.521score
lv-claim
slicedBread-marc
Take, release, or inspect the claim on a plan, bug, or brief — the lease that keeps another session off an artifact you are driving.
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 lv-brainstorm 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
lv-brainstorm skill. Agent Mart. https://agentmart.dev/skills/slicedbread-marc-lv-brainstorm (last updated 2026-10-24)