Skip to content

ucsahinn/agentchef / plugins/agentchef/skills/agent-brief

agent-brief

Verified

Write the task brief an orchestrator hands to another agent (subagent, Codex worker, or peer session) and check the handoff that comes back. Use before any delegation that has a goal, a write scope, or acceptance criteria; skip it for a one-line read-only question.

references/brief-template.md

View on GitHub
# Brief Template

Copy, fill every field, and run `brief-check` before sending. Labels may be
English, Turkish, or both ("Goal / Hedef").

```markdown
## Goal
Make new Claude sessions start fewer MCP processes.

## Evidence
- `claude mcp list` peak: 35 processes, 1919 MB commit (measured 2026-10-04).
- `~/.claude.json` user scope starts playwright and chrome-devtools in every session.

## Write scope
Repository agentchef: `catalog/mcp-servers.json`, `templates/codex/config.*.toml`, `docs/mcp-catalog*.md`.

## Boundaries
- Do not edit `~/.claude.json` or any live config; that is a later approved step.
- Do not touch the Beyin vault.

## Done when
- `npm run validate` passes.
- The plugin's `mcp/claude.mcp.json` lists only context7 and serena.

## Return format
Outcome, Evidence, Changed scope, Risks, Open questions, Next verification.

## User's words
"pc çok yavaşladı ... canlı ve kullanılan hiçbirşeye dokunmadan hızlandır"

## Environment traps (optional)
none

## Isolation (optional)
none
```

## Handoff Template

```markdown
Outcome: the plugin ships context7 and serena only.
Evidence: `node --test scripts/tests/mcp-launch.test.mjs` 4/4; fast 34/34.
Changed scope: catalog/mcp-servers.json, templates/codex/config.unix.toml, ...
Risks: a project that relied on the global Playwright must add it per project.
Open questions: none.
Next verification: open a new Claude session and run `claude mcp list`.
Verdict: VERIFIED
Mutation check: catalog/mcp-servers.json; templates/codex/config.unix.toml
```
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall