Skip to content
Agentmart
Submit a tool

health

Run a Brain Check — audit your second brain across Setup, Connections, Beads, Memory, Rules, and Inbox.

Skill by hamchowderr

IndexedMIT★ 2
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/hamchowderr-health

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Run a Brain Check — audit your second brain across Setup, Connections, Beads, Memory, Rules, and Inbox. Use when user says "brain check", "health check", "system status", "what's broken", "what needs attention", "check connections", "what's backed up".

Use cases: security

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Python package

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/health
  • Your home directory (~)

Tools it pre-approves 4

From allowed-tools in SKILL.md: the agent may use these without asking you each time.

  • Bash(python3:*),
  • Bash(node:*),
  • Bash(curl:*),
  • Bash(braynee-self-test:*)

What's inside

Skill name
health
Description
Run a Brain Check — audit your second brain across Setup, Connections, Beads, Memory, Rules, and Inbox. Use when user says "brain check", "health check", "system status", "what's broken", "what needs attention", "check connections", "what's backed up".
License
MIT
Allowed tools
Bash(python3:*),, Bash(node:*),, Bash(curl:*),, Bash(braynee-self-test:*)
Instructions
87 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts
  • Pre-approves some shell commands

Files (4)

  • SKILL.md
  • health.py · scripts/
  • rules_check.py · scripts/
  • test_rules_check.py · scripts/

SKILL.md (first lines)

# Health Skill — Brain Check

Audits the Braynee system across six areas and surfaces what's broken or backed up right now.

## Commands

```bash
# Full Brain Check (all six areas)
python3 {baseDir}/scripts/health.py check

# Check tools and plugin scripts are installed
python3 {baseDir}/scripts/health.py setup

# Verify live integrations are reachable
python3 {baseDir}/scripts/health.py connections

# Beads repo health — `bd doctor` + secret/gitignore drift, plus traceability
# hygiene (missing sections, open-but-landed, stale, create-time guard)
python3 {baseDir}/scripts/health.py beads

# Check Claude's context is current
python3 {baseDir}/scripts/health.py memory

# Check the instruction layers: budgets, dead references, skipped AGENTS.md, copies
python3 {baseDir}/scripts/health.py rules [--verbose]

# Surface what's backed up or unprocessed
python3 {baseDir}/scripts/health.py inbox

# Plugin self-test — validates every hook, monitor, script, skill, agent
node {baseDir}/../../bin/braynee-self-test
```

## Self-Test (`self-test` subcommand)

Run when something feels broken with braynee itself, or after editing the plugin source. Validates:

- All 21 hook scripts parse and execute with mock stdin (no crashes)
- hooks.json + monitors.json + plugin.json schema valid
- All 17 skills + 6 agents have valid frontmatter
- All bundled scripts (qmd-wrapper, vault-query, beads-dashboard) are callable
- All 3 monitors can boot

Exit code 0 = everything passed. Non-zero = at least one check failed (output shows which).

For machine-readable output use `--json`:
```bash
node {baseDir}/../../bin/braynee-self-test --json
```

The same self-test runs in CI on every push to main across Ubuntu, macOS, and Windows.

## Brain Check Framework

**Setup** — Are the tools and scripts installed and working?
- `obsidian`, `bd`, `node`, `python3`, `qmd` all available?
- Plugin scripts reachable via `{baseDir}`?

**Connections** — Are live integrations actually reachable?
- Beads (issue tracking)
- ProtonMail CLI
- QMD index

**Beads** — Is the repo's issue tracker healthy, traceable, and free of committed secrets? (runs `bd doctor` in the current repo)
- No tracked secrets — e.g. a committed `.beads/.beads-credential-key` (AES-256 key)
- `.gitignore` carries the beads-required patterns (no drift)
- Surfaces secret/gitignore findings loudly **before** any repo goes public
- **Traceability hygiene** — issues missing required sections, implemented-but-still-open (`bd orphans`), stale issues, and the create-time validation guard (`validation.on-create`). The same checks the `beads-auditor` agent runs. The guard is the one thing this check **repairs**: when it is off (unset, `none`, or any value bd does not honour) it is set to `error`, which refuses an issue missing the sections its type needs and lets chores through. A repo that chose `warn` keeps it.

**Memory** — Is Claude loaded with current context?
- `~/.claude/CLAUDE.md` reflects current setup?
- Vault CLAUDE.md up to date?
- Priorities reflecting actual focus?

**Rules** — Are the instruction layers loading, within budget and consistent? Report-only; it never edits.
- User `CLAUDE.md` under 5 KB; always-on total (CLAUDE.md + rules without `paths:`) under 40 KB
- `paths:` globs are project-relative (an absolute glob never matches)
- No references to rule files, braynee skills or agents, or vault notes that don't exist, in your CLAUDE.md, rules, agents and braynee's shipped text
- Repos under `BRAYNEE_PROJECTS_DIR` (default `~/code`): AGENTS.md that Claude Code skips because CLAUDE.md lacks `@AGENTS.md`, CLAUDE.md identical to AGENTS.md, CLAUDE.md copied between repos, instruction files over 5 KB
- MEMORY.md under 200 lines with entries of 150 characters or fewer; feedback memories and untyped memories naming a rule (a correction belongs in the rule's own file)

**Inbox** — What's backed up or unprocessed right now?
- Inbox items waiting in `Inbox/`
- Sessions not yet synced to vault
- Stale open tasks

From hamchowderr/braynee, MIT.

More from hamchowderr/braynee

Signals

Stars
2
Forks
0
Open issues
1
Last commit (this folder)
Oct 20, 2026 (8 days ago)
Last release
Oct 12, 2026 (16 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from hamchowderr

Cite this page

health skill. Agent Mart. https://agentmart.dev/skills/hamchowderr-health (last updated 2026-10-24)