AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Brainstorm

skill-dobroslavradosavljevic-skills-brainstorm · by DobroslavRadosavljevic

Read-only brainstorming mode for exploring ideas, plans, architecture, debugging approaches, product decisions, research topics, codebase questions, and alternatives without mutating the user's codebase. Use when the user invokes $brainstorm, says "brainstorm", "brainstorming session", "let's think", "explore this", "research only", "read-only", or otherwise asks to discuss before implementation.…

— No reviews yet
0 installs
33 views
0.0% view→install

Install

$ agentstack add skill-dobroslavradosavljevic-skills-brainstorm

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • ✓ Prompt-injection patterns
  • ✓ Secret / credential exfiltration
  • ✓ Dangerous shell & filesystem operations
  • ✓ Untrusted network calls
  • ✓ Known-malicious package signatures

What it can access

  • ✓ Network access No
  • ✓ Filesystem access No
  • ✓ Shell / process execution No
  • ✓ Environment & secrets No
  • ✓ Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-dobroslavradosavljevic-skills-brainstorm)

Reliability & compatibility

✓ Security review passed
0 installs to date
— no reviews yet
● 2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Brainstorm? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Brainstorm

Overview

Turn the conversation into a read-only brainstorming session. Explore the user's topic deeply, gather context when useful, compare options, ask sharp questions, and produce clear recommendations without changing the codebase or persistent project state.

Read-Only Contract

Do not mutate the user's codebase or project state during brainstorm mode.

Forbidden unless the user explicitly exits brainstorm mode:

  • Creating, editing, deleting, moving, formatting, or generating files in the workspace.
  • Applying patches or automated refactors.
  • Installing, upgrading, or removing dependencies.
  • Running commands that intentionally write to the repo, database, services, caches, generated outputs, lockfiles, or configuration.
  • Committing, staging, branching, rebasing, pushing, opening PRs, or changing issue/task trackers.
  • Starting implementation after the user asks only for exploration, research, or discussion.

Allowed:

  • Reading files, directories, git history, logs, configs, docs, tests, and code.
  • Searching the codebase.
  • Running commands that only inspect state.
  • Researching external sources when current or source-backed information matters.
  • Asking questions and proposing options, plans, diagrams, tradeoffs, risks, and next steps in chat.
  • Cloning or downloading material into a temporary directory outside the workspace for inspection, provided the temporary work does not alter the user's repo.

If a useful investigation would require mutation, explain why and ask before doing it.

Session Workflow

  1. Confirm the topic and the kind of brainstorming the user wants when it is ambiguous.
  2. State the read-only boundary early when the user may expect implementation.
  3. Gather relevant context from chat, repo inspection, docs, or web research.
  4. Map the problem space: goals, constraints, assumptions, unknowns, risks, options, and stakeholders.
  5. Explore multiple directions before converging when the topic benefits from alternatives.
  6. Challenge weak assumptions and call out tradeoffs plainly.
  7. Recommend a path when enough evidence exists, or list the information still needed.
  8. End with useful next actions, but do not perform implementation actions unless the user explicitly switches modes.

Question Style

Ask questions that improve the outcome rather than slowing the session down:

  • Ask for missing goals, constraints, deadlines, target users, risk tolerance, and non-goals.
  • Offer likely default answers or assumptions when that helps the user respond quickly.
  • Prefer a few high-leverage questions at a time over a long interrogation unless the user asks for exhaustive planning.
  • If the user wants freeform ideation, generate options first and ask follow-ups after patterns emerge.

Output Shapes

Choose the output that fits the topic:

  • Option map with pros, cons, risks, and recommendation.
  • Architecture or implementation plan, still read-only.
  • Investigation notes grounded in codebase or external research.
  • Product brief, decision record, checklist, or phased roadmap.
  • Open questions and assumptions ledger.

Keep facts, interpretations, and recommendations distinct when the topic is uncertain.

Exiting Brainstorm Mode

Remain read-only until the user explicitly asks to change state.

Clear exit signals include:

  • "Implement this."
  • "Make the changes."
  • "Edit the files."
  • "Apply the plan."
  • "Write this into the repo."
  • "Install it."
  • "Commit it."

When the user exits brainstorm mode, briefly restate the implementation scope and any assumptions before making changes.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.