Use on the first user message of every new session — before any tool use, file reads, or repo exploration. Forces a brief understanding replay so the user can catch misinterpretations before execution begins.
From dev-skillsnpx claudepluginhub igor1309/skills --plugin dev-skillsThis skill uses the workspace's default tool permissions.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Executes pre-written implementation plans: critically reviews, follows bite-sized steps exactly, runs verifications, tracks progress with checkpoints, uses git worktrees, stops on blockers.
Announce: "Pausing to confirm I understand before diving in."
Without using any tools, formulate a 2-3 sentence summary of the user's request. Be specific — name files, modules, behaviors if mentioned. Flag anything ambiguous.
Output this block:
## Cold Start Checkpoint
**My understanding:** [2-3 sentence summary]
**Assumptions I'm making:**
- [assumption 1]
- [assumption 2]
- ...
**What would you like to do?**
1. Go — understanding is correct, proceed
2. Clarify — let me ask a few questions first
3. Restate — let me re-explain what I need
Keep it brutally short. No preamble, no meta-commentary. 3-5 assumptions max — only ones that would derail execution if wrong.
For crystal-clear, single-scoped prompts (e.g., "bump version to 2.1.0 in Package.swift"): one-sentence summary, skip assumptions, just show the Go/Clarify/Restate options.