- Home
- Skills
- Research & Analysis
- tt brainstorm
Works with the AI Agents you already use
tt brainstorm
Multi-lens research engine that generates sourced ideas, feasibility checks, and contrarian analysis for complex topics.
$7
/moSubscription · cancel anytime
Secure checkout via Stripe
- Always the latest version
- Delivered live to your agent via MCP
- Cancel anytime, access ends at period end
tt brainstorm
Example session with this skill installed
/tt-brainstorm topic="migrating our monolithic auth to a distributed service" depth=deep lenses=feasibility,contrarian
- Read your context and instructions
- Compiled the tt brainstorm
- Generated the document
I have completed a deep-dive brainstorm on migrating the monolithic authentication system. I analyzed the topic through feasibility and contrarian lenses over two rounds of research.
Brainstorm Dossier: Distributed Auth Migration
| Idea ID | Concept | Type | Source/Evidence |
|---|---|---|---|
| AUTH-01 | Sidecar-based JWT validation | Derived | Based on current Kubernetes ingress patterns in infra/ |
| AUTH-02 | Centralized Shared Session Store | Found | Documented in legacy-auth-specs.pdf as a prerequisite |
| AUTH-03 | Incremental Proxy Steering | Derived | Inferred from current Nginx configuration limits |
Feasibility Analysis
The primary constraint is the legacy session cookie format which is encrypted using a provider-specific library. A distributed service would require either a shared decryption key or a shim layer. Technical feasibility is high for JWT adoption but medium for full session parity without a refactor of the UserSession class.
Contrarian Perspective
Latency Penalty: Introducing a remote auth check adds 15-40ms per request. For high-frequency internal APIs, this may violate existing SLAs.
Complexity Risk: Moving to a distributed model introduces "split-brain" risks during network partitions that the current monolith avoids by localized state.
Security Surface: A new service endpoint increases the internal attack surface and requires fresh mTLS configuration.
Assumptions & Open Questions
- Does the mobile app support Bearer tokens, or is it hardcoded for cookies?
- Can we tolerate a 20ms increase in P99 latency for the dashboard?
Next steps
- Review the generated dossier at
docs/brainstorms/2023-10-27-auth-migration.md. - Run a latency simulation on the proposed sidecar architecture.
- Decide if the contrarian latency risks outweigh the scalability benefits.
legacy-auth-specs.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Subscription · always the latest version
- Secure checkout via Stripe
- Cancel anytime
- Delivered live via MCP
About this skill
Vague project requirements and shallow "top 5" lists often lead to wasted development cycles. This skill forces deep, multi-perspective research into a topic before any code is written, ensuring you understand the market, feasibility, and contrarian risks.
What it does
- Multi-lens research executes independent analysis across problem, market, technical feasibility, and contrarian perspectives.
- Sourced idea generation returns concepts categorized as either cited from existing data or derived through logical inference.
- Conflict preservation maintains differing judgments side-by-side rather than forcing a biased consensus, allowing for manual review.
- Automated dossier creation writes a structured report to
docs/brainstorms/including source ledgers, assumptions, and discarded paths. - Subagent orchestration manages read-only research workers to parallelize deep-dive tasks without exceeding token or concurrency caps.
How it works
- Initialize the topic and define parameters like depth (quick, standard, deep) and specific lenses.
- Clarify the goal and constraints through up to four targeted questions to narrow the search space.
- Execute research rounds using internal subagents to pull from supplied sources, repository context, and authorized memory.
- Synthesize findings by deduplicating claims and mapping them to stable idea IDs for traceability.
Frameworks & tools
This skill is framework-agnostic but integrates with local file systems for documentation. It works best in environments supporting subagent delegation like Claude Code or Cursor.
Why this beats prompting it yourself
Generic prompts often suffer from confirmation bias or shallow synthesis. This skill enforces a structured "think tank" methodology that explicitly seeks out contrarian viewpoints and tracks the provenance of every claim, preventing the agent from hallucinating "good ideas" without evidence.
Use cases
- Validating the technical feasibility of a new microservice architecture.
- Analyzing market competitors before building a specific feature.
- Stress-testing a product direction with a dedicated contrarian lens.
- Gathering alternative implementation options for a legacy refactor.
Known limitations
Does not perform external writes, account logins, or production changes. Limited to a maximum of three research rounds and eight total workers.
Read-only towards outside systems. Max 3 rounds and 8 workers. Requires file write authorization for dossier storage. Does not perform implementation or code delivery.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 27 days ago