# briefing

A plugin that reads the skills a subagent declares and puts them into its context before the agent starts working.

- plugin · `type=plugin` · Claude Code · ★ 5
- Author: Getty
- License: Artistic-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/Getty/briefing
- Updated: 2026-10-06
- HTML: https://affpapa.org/en/nevibecode/plugins/briefing

## What it does

Subagents begin with a clean context window, and skills load lazily: what sits there is a list of names and descriptions. An agent spawned for a specific piece of expertise still has to notice it needs that expertise. Sometimes it does, sometimes it works uninformed, and the output looks the same either way.

briefing removes the guesswork. At spawn time a hook reads the declaration, resolves each skill and injects the full text. In Claude the hook fires on `PreToolUse` for the `Agent` tool, and the declaration lives in a `briefing:` block in the frontmatter of `.claude/agents/*.md`. Each skill arrives inside its own `<skill name=... dir=...>` element; `dir` points at the folder, so `references/` and `scripts/` stay on disk and load on demand.

A declared skill that cannot be resolved stops the spawn — the plugin will not run on a partial briefing. Run `briefing-doctor` to check declarations ahead of time: it walks every agent in the project and exits 1 if any spawn would fail, which makes it usable in CI. Skill names resolve the same way everywhere. A bare name searches project skills, then user skills, then installed plugins; a namespaced one such as `superpowers:brainstorming` goes straight to that plugin. A marketplace plugin that is off in `enabledPlugins` resolves to nothing.

## Install

### Claude Code (native platform)

**Plugin marketplace**
In the agent chat (not a shell):
```text
/plugin marketplace add anthropics/claude-plugins-community
```
In the agent chat (not a shell):
```text
/plugin install briefing@claude-community
```

## When to use it
- get the right skills into a subagent before it starts
- drop "load skill X first" instructions from the agent body
- catch a misspelled skill name before a spawn fails
- check every agent declaration in CI
- see which skills and how many kilobytes an agent received

## Example prompts
- Add a `briefing:` block with `getty-perl-core` and `superpowers:brainstorming` to `.claude/agents/reviewer.md`
- Run `briefing-doctor` and show me which agents would fail to spawn
- Why did the `backend` agent not receive the `kubernetes-rest` skill?

## Similar
- [daily-briefing](https://affpapa.org/en/nevibecode/skills/daily-briefing.md) — skill · 3300 installs
- [matter-briefing](https://affpapa.org/en/nevibecode/skills/matter-briefing.md) — skill · 608 installs
- [c-briefing](https://affpapa.org/en/nevibecode/skills/c-briefing.md) — skill · ★ 173
- [aa-executive-briefing](https://affpapa.org/en/nevibecode/skills/aa-executive-briefing.md) — skill · 329 installs
- [cja-executive-briefing](https://affpapa.org/en/nevibecode/skills/cja-executive-briefing.md) — skill · 324 installs
- [advertising-kickoff](https://affpapa.org/en/nevibecode/skills/advertising-kickoff.md) — skill · 20500 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md