Skills for Real Engineers. Straight from my .agents directory.
minutes-mcp-recall
Agent skill fromsilverstein/minutes
Route meeting-recall questions to the right Minutes MCP tool.
How to install minutes-mcp-recall
npx skills add silverstein/minutes# or add only this skill to Claude Code: copy the folder
tooling/skills/sources/minutes-mcp-recall/
# into
~/.claude/skills/minutes-mcp-recall/ (all projects)
.claude/skills/minutes-mcp-recall/ (one project)
1# /minutes-mcp-recall23Route a meeting-memory question to the smallest useful Minutes MCP tool. This4skill is tool-name-oriented for MCP hosts such as OpenClaw, Hermes-agent, and5other MCP clients. Use `minutes-search` instead when the host only has CLI or6shell access.78Read [the complete tool map](references/tool-map.md) only when the decision9table does not cover the request.1011## Decision table1213| User is asking... | Call |14|---|---|15| "What did we say about X?" or "Find that meeting about Y." | `search_meetings` |16| "What is the full transcript of this specific meeting?" | `get_meeting` after finding its id or date with `list_meetings` or `search_meetings` |17| About what is happening right now, mid-call | `read_live_transcript` with a cursor or time window |18| "What do I still owe?" or "Who owes me?" | `track_commitments` |19| "What is my history with this person?" | `get_person_profile` or `relationship_map` |20| "Have I contradicted myself?" or "Which decisions are stale?" | `consistency_report` |21| Broad research across many meetings | `research_topic` |2223## Tool-name prefix by host2425- OpenClaw and generic MCP clients use bare tool names such as26 `search_meetings`.27- Hermes-agent prefixes each tool with `mcp_minutes_`, so28 `search_meetings` becomes `mcp_minutes_search_meetings`.2930Apply the same prefix rule to every tool in the decision table and tool map.3132## Anti-patterns3334- Do not call `get_meeting` speculatively to browse. It returns a full35…
Excerpt shown under the repository's MIT license.
More skills in minutes
Similar skill repositories
An agentic skills framework & software development methodology that works.
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
Public repository for Agent Skills
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Production-grade engineering skills for AI coding agents.
