# Summarize Conversation Thread

> Use when catching up on a long GitHub issue or PR discussion thread, or asked to summarize a thread, extract key decisions, or list action items - produces a structured summary with TL;DR, decisions, action items, open questions, and next steps

- **Type:** Skill
- **Install:** `agentstack add skill-britt-claude-code-skills-summarize-conversation-thread`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [britt](https://agentstack.voostack.com/s/britt)
- **Installs:** 0
- **Category:** [Developer Tools](https://agentstack.voostack.com/c/developer-tools)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [britt](https://github.com/britt)
- **Source:** https://github.com/britt/claude-code-skills/tree/main/skills/summarize-conversation-thread
- **Website:** https://britt.github.io/claude-code-skills/

## Install

```sh
agentstack add skill-britt-claude-code-skills-summarize-conversation-thread
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Summarize Conversation Thread Skill

Read long GitHub issue or PR threads and produce concise summaries highlighting what matters.

## When to Use

Activate when:
- Catching up on a lengthy discussion
- Preparing to contribute to an existing thread
- Extracting decisions from a closed issue
- Identifying action items from a discussion

**When NOT to use:** Skip for short threads (fewer than ~5 comments) — just read them directly.

## Getting the Thread

First, obtain the full thread content:

- **Issues**: `gh issue view  --comments` (requires the `gh` CLI, authenticated)
- **PRs**: `gh pr view  --comments`
- **No gh CLI or no repo access**: Ask the user to paste the thread content

## Output Structure

```markdown
## Summary: [Issue/PR Title]

**Status**: [Open/Closed] | **Comments**: [X] | **Last Activity**: [date]

### TL;DR

[2-3 sentence summary of the core discussion]

### Key Decisions

- **[Decision]**: [context and who decided]

### Action Items

- [ ] [Action] — @[owner]

### Open Questions

- [Unresolved question needing input]

### Next Steps

[What should happen next]
```

## Guidelines

### What to Include

- Decisions made and who made them
- Action items with clear owners
- Unresolved questions or blockers
- Changes in direction or scope
- Links to key comments

### What to Skip

- Redundant "+1" or "me too" comments
- Off-topic tangents
- Superseded proposals
- Resolved sub-discussions

### Handling Edge Cases

- **Stale threads** (>30 days inactive): Note this and flag if decisions may be outdated
- **Heated discussions**: Present all perspectives neutrally
- **No resolution**: Clearly state "No decision reached" and summarize options discussed
- **Long threads** (50+ comments): Break into phases if discussion evolved significantly

## Example

```markdown
## Summary: Add dark mode support (#142)

**Status**: Open | **Comments**: 23 | **Last Activity**: 2 days ago

### TL;DR

Team agreed to implement dark mode using CSS variables. Debate on whether to auto-detect system preference or default to light mode resolved in favor of auto-detect.

### Key Decisions

- **Use CSS variables**: Easier to maintain than separate stylesheets — @sarah
- **Auto-detect system preference**: Better UX, users can override — @mike

### Action Items

- [ ] Create color token system — @sarah
- [ ] Add toggle to settings page — @alex

### Next Steps

Sarah to submit initial PR with color variables by Friday.
```

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [britt](https://github.com/britt)
- **Source:** [britt/claude-code-skills](https://github.com/britt/claude-code-skills)
- **License:** MIT
- **Homepage:** https://britt.github.io/claude-code-skills/

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-britt-claude-code-skills-summarize-conversation-thread
- Seller: https://agentstack.voostack.com/s/britt
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.