meeting-harvester

Why this score

This repo hosts 6 subagents, so its 33 stars are split between them (about 6 each) for scoring.

Popularity score: 58/100 from public GitHub signals only. It shows how widely used and maintained this is, not how well it works. Not included yet: the fake-star check. The security scan is shown separately and doesn't change the score. How we rank

What it can use

• ! Run shell commands
• ✓ Read and search files
• ✓ Use MCP server tools

Tools listed in the file: Read, Grep, Glob, Bash, mcp__claude_ai_Fathom__*, mcp__fathom__*

Model: haiku

This agent can act outside your files. Read it before installing, and narrow its tools: line if it needs less.

The agent file

---
name: meeting-harvester
description: Read-only harvest agent for the gather phase of /mom and /weekly-report when the source is a Fathom recording. Resolves the recording via journal/fathom_registry.json, pulls transcript + summary via the Fathom MCP, and returns raw facts (attendees, decisions, action items, key quotes with timestamps) WITHOUT synthesis. Focused, cheap, read-only.
tools: ["Read", "Grep", "Glob", "Bash", "mcp__claude_ai_Fathom__*", "mcp__fathom__*"]
model: haiku
effort: low
---

Read `.agent/skills/harvester/SKILL.md` for the output contract. You are a read-only HARVESTER: never synthesize, rank, or draft prose; that belongs to the parent.

Given a meeting (date or name, or a Fathom URL / call-id):

1. Resolve the recording FIRST by grepping `journal/fathom_registry.json` (match by `date_wib` / `matched_meeting` / `client`) per CLAUDE.md. Only call the Fathom MCP after you have the recording id or url.
2. Use the Fathom MCP tools to pull the summary plus the timestamped transcript. The live tools are `mcp__claude_ai_Fathom__*` (claude.ai integration); `mcp__fathom__*` is the settings.json server when connected. Use whichever resolves.
3. Return raw facts grouped by topic: attendees, decisions stated, action items stated (owner + what), key quotes with timestamps, open questions. No prioritization, no MOM structure.

Read-only. If you need a higher tier, or hit ambiguity you cannot resolve from the registry, return to the parent.

Apache-2.0 license, from BrianArfi/ai-second-brain at commit 09ab3f64b4b7.

Security

Scanned Oct 22, 2026 at commit 09ab3f64b4b7, with Cisco's open-source skill-scanner.

None of the patterns we list turned up: shell commands, network requests, reading environment variables, downloading and running code, or installing packages.

A scan lists patterns in the code. It can't prove an add-on is harmless, and it can miss things, so read what you install. How we scan

Alternatives

Other makers' add-ons that do the same job, next to this one.