The request "coldstart" is ambiguous: it could mean starting an AI session without context, a cold-start interview, or a cold-start strategy. Without a named product or platform, the closest matches are DeerDawn, which gives AI sessions a standing project brief so they never start cold 2, and Anthropic's cold-start-interview action for learning a team's practice profile on first use 1.
A question back: Do you mean starting an AI session with no prior context, a cold-start interview or onboarding flow, or cold-start strategy for a product launch, and which platform or tool are you using?
- 1
DeerDawndeerdawn.comverified and signed
Gives Claude, ChatGPT, Cursor and Codex a standing brief of shipped, decided and open work before each session, so the AI resumes instead of starting cold 2, via the start_session action that returns current task, recent decisions and open questions 12.
official:MCP server · @deerdawn/mcp-serverMCP server · com.deerdawn/deerdawn+1 community tools built for it
for agents: Load project brief · Load project brief · start_session
- 2
Sources
- 1Home \ Anthropicanthropic.com · read 2026-10-29
- 2Never start cold · DeerDawndeerdawn.com · read 2026-10-29
- 11cold-start-interview: declares "cold-start-interview" (the record we read and signed)unfragile.ai · read 2026-10-24
- 12DeerDawn: declares "Load project brief" (the record we read and signed)unfragile.ai · read 2026-10-24
For agents: the 10 actions behind this answer
cold-start-interview · cold-start-interview verified and signed
Run the cold-start interview to learn your commercial contracts practice and write your team practice profile. Use on first use of the plugin, when `~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md` is missing or still contains template placeholders, or when the user says "set up
Load project brief · DeerDawn verified and signed
Return the current DeerDawn brief for a project — active task, recent decisions, tech stack, open questions, and where the last session left off — so a calling agent starts warm instead of cold. e.g. Load the brief for my DeerDawn project. e.g. What is the current state of the checkout-rewrite proje
gtm-0-to-1-launch · gtm-0-to-1-launch verified and signed
Launch new products from idea to first customers. Use when launching products, finding early adopters, building launch week playbooks, diagnosing why adoption stalls, or learning that press coverage does not equal growth. Includes the three-layer diagnosis, the 2-week experiment cycle, and the launc
docker-destructive-guardrails · docker-destructive-guardrails verified and signed
Use this skill before running, or recommending, any Docker command that deletes, wipes, resets, or otherwise irreversibly changes state — even if the user just says to "clean up", "clear the cache", "start fresh", "wipe everything", "nuke it", "reset", "force remove", or "tear down" Docker resources
create_task · supabase found in code, not yet checked
Create an open task for yourself and return it. Each call creates a new task; do not retry blindly after a connection failure.
notion · @notionhq/client verified and signed
· Playwright website read, tools not yet checked
Web automation and testing for apps, scripts, and AI agents
begin_repro_run · shin-watcher-oss found in code, not yet checked
Call this FIRST before any repro work. Creates an isolated run directory, allocates a proxy port, and registers the run with the dashboard.
worker-tail-logs-start-tail · Cloudflare API verified and signed
Start Worker Tail
· Ollama website read, tools not yet checked
Ollama is the easiest way to run and build with open AI models locally or in the cloud on macOS, Windows, and Linux.
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.