Coaching Session Summarizer

Community

glebis

coaching-session-summarizer

This skill should be used to summarize coaching or therapy session transcripts after a Fathom/Granola sync. The agent analyzes the transcript itself (no API key, runs on the subscription) and appends key insights, decisions, action items, and trail connections. Supports quick extraction or deep analysis with cross-session pattern detection.

Overview

Publisher	glebis
Repository	`claude-skills`
Skill name	`coaching-session-summarizer`
Stars	379
Forks	56
Bundled files	2
License	MIT
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
2 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by glebis on GitHub. Read the source before you install it.

Installation

Install the Coaching Session Summarizer AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/glebis/claude-skills.git /tmp/claude-skills
mkdir -p .claude/skills
cp -r /tmp/claude-skills/coaching-session-summarizer .claude/skills/coaching-session-summarizer

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Coaching Session Summarizer in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Coaching Session Summarizer on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Coaching Session Summarizer is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Coaching Session Summarizer

Overview

Analyzes a coaching/therapy session transcript and appends a structured summary (key insights, decisions, action items, deep analysis, connected trails) to the note.

The agent (Claude Code) performs the analysis directly — reading the transcript and writing the summary in this session. There is no Anthropic API call and no billing; it runs entirely on the active subscription. A legacy API-based script is kept only as a headless fallback (see bottom).

When to Use This Skill

A new Fathom/Granola transcript was synced to the vault (coaching or therapy)
User asks to summarize/analyze a session (/summarize-session [file] or similar)
After calendar-sync or a Granola export, when a new *-coaching.md, *-therapy.md, or *-session.md file appears — offer to summarize it

Workflow (agent-driven — default)

Do this in-session with native tools. No API key required.

Step 1 — Gather context

Run the deterministic helper to get the transcript text, previous sessions, and the trail list in one shot:

bash
python3 ~/.claude/skills/coaching-session-summarizer/scripts/gather_context.py \
  <transcript-file> --vault ~/Brains/brain

It prints:

Previous sessions with the same participant (paths) — Read these only in deep mode, for cross-session pattern detection
Available trails — pick 2–4 most relevant to link
Session content — the summary + transcript to analyze (any prior AI-Generated Summary is stripped so re-runs stay clean)

Pass --participant <name-slug> if the filename doesn't encode the person (e.g. Granola exports titled by topic): --participant gleb-kalinin.

Step 2 — Analyze

Read the session content and extract, in the analytical voice of a session analyst (objective, using the speaker's authentic language where it matters):

Key Insights — 3–5 main realizations / breakthroughs / observations
Decisions Made — concrete choices or commitments
Action Items — specific next steps; prefix time-sensitive ones with [URGENT] and scheduling items with [SCHEDULING]
Session Themes — 2–3 recurring topics or patterns

Deep mode (default for therapy and milestone sessions) — also Read the previous sessions and add:

Pattern Detection — themes recurring across sessions
Progress Assessment — movement on earlier commitments
Energy/Motivation Markers — shifts in energy, resistance, affect
Potential Obstacles — what might block progress

Step 3 — Append with Edit

Append the summary to the end of the transcript file using Edit (never overwrite existing content). Match this exact structure:

markdown
## AI-Generated Summary

*Generated: YYYY-MM-DD*

### Key Insights
- ...

### Decisions Made
- ...

### Action Items
- [URGENT] ...
- ...

### Session Themes
- ...

## Deep Analysis

- **Pattern Detection**: ...
- **Progress Assessment**: ...
- **Energy/Motivation Markers**: ...
- **Potential Obstacles**: ...

## Connected Trails

- [[Trails/Trail - <Name>|<Name>]]
- [[Trails/Trail - <Name>|<Name>]]

Use the current date (date +%Y-%m-%d) in the Generated line. Omit the Deep Analysis section in quick mode. Verify trail link names against the printed trail list — case and exact wording matter for Obsidian links.

Modes

quick — Key Insights, Decisions, Action Items, Themes. Skip Deep Analysis and previous-session reads.
deep (recommended for therapy / milestones) — everything, including reading previous sessions for pattern detection.

Integration with Sync

After calendar-sync or a Granola/Fathom export, check for new session files (*-coaching.md, *-therapy.md, *-session.md). If one appears, offer: "New session detected — summarize now?" Default to deep mode for therapy.

Notes

Preserves the original transcript intact; the summary is always appended.
Trail linking requires the Trails/ directory in the vault root.
Cross-session comparison works best with consistent naming: YYYYMMDD-name-coaching.md / YYYYMMDD-name-therapy.md.
Re-running is safe: gather_context.py strips any prior AI-Generated Summary before printing, so the agent analyzes only the raw session. (Delete the old ## AI-Generated Summary block from the file before re-appending if you want to replace rather than stack summaries.)

Resources

scripts/

gather_context.py — (default path) deterministic context gatherer, no API. Prints transcript text + previous sessions + trail list for the agent to analyze in-session.
summarize_session.py — legacy / headless fallback. Calls the Anthropic API directly (model via SUMMARIZER_MODEL, default claude-sonnet-4-6) and bills a funded ANTHROPIC_API_KEY. Use only when no interactive agent is available (e.g. cron). Exits with a clear message if the key has no credit.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

scripts/gather_context.py3 KB
scripts/summarize_session.py11 KB

Frequently asked questions

What does the Coaching Session Summarizer AI skill do?

Why use Coaching Session Summarizer on TypingMind?

Because you install it once and use it with any model. Coaching Session Summarizer is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Coaching Session Summarizer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/glebis/claude-skills/tree/main/coaching-session-summarizer. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Coaching Session Summarizer?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Coaching Session Summarizer?

As many as you like. As long as a model supports skills, you can use Coaching Session Summarizer with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Coaching Session Summarizer AI skill free?

Yes. It is published on GitHub by glebis under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Loop Me

mattpocock

CommunityPopular

Grill me about specs for the workflows I want to build, within this workspace.

Scaffold Exercises

Create exercise directory structures with sections, problems, solutions, and explainers that pass linting. Use when user wants to scaffold exercises, create exercise stubs, or set up a new course…

Api Design

Patrones de diseño REST API incluyendo nomenclatura de recursos, códigos de estado, paginación, filtrado, respuestas de error, versionado y rate limiting para APIs de producción.

Backend Patterns

Patrones de arquitectura backend, diseño de API, optimización de base de datos y buenas prácticas del lado del servidor para Node.js, Express y rutas API de Next.js.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

Coaching Session Summarizer

Overview

When to Use This Skill

Workflow (agent-driven — default)

Step 1 — Gather context

Step 2 — Analyze

Step 3 — Append with Edit

Modes

Integration with Sync

Notes

Resources

scripts/

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now