Chalk Labs
Need help deploying onboard?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
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/danielrosehill-onboard
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.
About
Use cases: backend, security
Reviews
Write a reviewNo 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. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/onboard
What's inside
- Skill name
onboard- Description
- First-run setup for the media-assistant-ops plugin. Capture the user's local Music Assistant deployment URL + API key, snapshot the current speaker/player roster for future context, vet the overall setup (providers, connectivity, auth), and summarise the library (track/album/artist counts, active music providers). Use when the plugin is freshly installed, when `config.json` is missing, or when the user wants to re-baseline against a new server.
- License
- MIT
- Instructions
- 99 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Onboard — Music Assistant
End-to-end first-run: capture credentials, verify the server responds, snapshot the environment, and produce a context file the other skills can rely on.
## Phase 1: Capture credentials
1. Resolve the data root and ensure directories exist:
```bash
DATA_ROOT="${CLAUDE_USER_DATA:-${XDG_DATA_HOME:-$HOME/.local/share}/claude-plugins}/media-assistant-ops"
mkdir -p "$DATA_ROOT/data" "$DATA_ROOT/state"
```
2. If `$DATA_ROOT/config.json` exists, display the current values (redact API key to last 4 chars) and confirm the user wants to overwrite.
3. Ask the user for:
- **Local deployment URL** — full base URL (e.g. `http://10.0.0.50:8095`). Accept an IP alone and assume `http://` + default port `8095`.
- **WAN URL / hostname** (optional).
- **API key** (required) — long-lived token from Music Assistant → Settings → Security.
4. Write `$DATA_ROOT/config.json` with `lan_url`, `wan_url`, `api_key`. `chmod 600` the file.
Never echo the full API key back — only the last 4 characters for confirmation.
## Phase 2: Vet the setup
Run these checks against the LAN URL (fall back to WAN if LAN is unreachable and a WAN URL was provided). Report each as ✓ / ✗ with a short note.
1. **Reachability** — GET `/api/info` with the Bearer token. Capture server version, Python version, OS.
2. **Auth** — confirm the API key was accepted (200 vs 401/403).
3. **Provider health** — GET `/api/providers`. For each provider, report `name`, `type` (music / player / metadata / plugin), and `available` status. Flag any provider in an error state.
4. **Port / firewall** — if GET `/api/info` fails, attempt a raw TCP connect to the port and report whether the service is up vs whether it's an auth problem.
Produce a short **Setup Vetting Report** the user can read at a glance.
## Phase 3: Snapshot the speaker/player roster
1. GET `/api/players` and persist the full response to:
```
$DATA_ROOT/data/players-snapshot.json
```
2. Write a human-readable summary to `$DATA_ROOT/data/players-snapshot.md` with one row per player:
| Name | Player ID | Type | Provider | Default volume | Group membership |
This file is the future-context artefact — other skills can grep it to resolve names → IDs and it gives you (Claude) a cheap way to know what hardware exists without a live call.
3. If any player is part of a sync group, note the group and which player is the leader.
## Phase 4: Library summary
1. Hit the library summary endpoints (if the running MA version exposes them) and persist raw JSON to `$DATA_ROOT/data/library-snapshot.json`:
- GET `/api/library/tracks?limit=1` — read total count from pagination meta
- GET `/api/library/albums?limit=1`
- GET `/api/library/artists?limit=1`
- GET `/api/library/playlists?limit=1`
- GET `/api/library/radios?limit=1`
If a given endpoint 404s on this MA version, skip it and note "not supported on this server version" in the report — don't fail the whole onboarding.
2. List active **music providers** (from Phase 2's provider fetch, filtered to `type=music`) with track count per provider if the API exposes it.
3. Write a human-readable `$DATA_ROOT/data/library-snapshot.md`:
```
# Music Assistant Library — snapshot <ISO date>
- Tracks: N
- Albums: N
- Artists: N
- Playlists: N
- Radios: N
## Music providers
- spotify (connected) — 12,345 tracks
- plex (connected) — 4,567 tracks
- filesystem (connected) — 890 tracks
```
## Phase 5: Final report to the user
Print a single consolidated summary covering:
- ✓/✗ for each vetting check
- Server version + address used
- Player count and any notable states (offline, grouped, DSP-capable)
- Library totals and active music providers
- Paths of the snapshot files written under `$DATA_ROOT/data/`
- Next suggested commands: `/media-assistant-ops:status`, `/media-assistant-ops:list-players`
## Do not
- Do not write anything to `~/.claude/` or inside the plugin install directory.
- Do not echo the full API key after capture.
- Do not block onboarding if the library endpoints are unsupported — degrade gracefully with a note.
- Do not overwrite an existing `config.json` without explicit user confirmation.From danielrosehill/Claude-Media-Assistant-Ops-Plugin, MIT.
More from danielrosehill/Claude-Media-Assistant-Ops-Plugin
- 0.322score
save-dsp-settings
danielrosehill
Snapshot a Music Assistant player's current DSP (EQ bands + filters) configuration and persist it as a named preset under $CLAUDE_USER_DATA.
SkillDFreshness: DClaude Code +9 - 0.322score
control-player
danielrosehill
Send a playback command to a Music Assistant player — play, pause, stop, next, previous, or set volume.
SkillDFreshness: DClaude Code +9 - 0.322score
update-dsp-settings
danielrosehill
Apply a DSP configuration to a Music Assistant player — either by loading a previously saved preset from disk or by adjusting specific EQ bands/filters…
SkillDFreshness: DClaude Code +9 - 0.322score
apply-podcast-preset
danielrosehill
Apply a recommended parametric-EQ DSP preset tuned for spoken-word / podcast listening to a Music Assistant player.
SkillDFreshness: DClaude Code +9 - 0.320score
reference
danielrosehill
Shared reference for the Music Assistant local API — base URL, auth, the WebSocket command bus, the DSP/EQ schema, and the data-storage layout this plugin uses.
SkillDFreshness: DClaude Code +9 - 0.318score
status
danielrosehill
Verify the plugin can reach the configured Music Assistant server.
SkillDFreshness: DClaude Code +9 - 0.318score
list-players
danielrosehill
Enumerate all players known to the Music Assistant server along with their current state (playing/paused/idle), volume, and currently-playing track if any.
SkillDFreshness: DClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Apr 21, 2026 (6 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from danielrosehill
- 0.470score
plugin-data-storage
danielrosehill
Canonical reference for where Claude Code plugins should persist user data.
SkillAFreshness: A★ 2Claude Code +9 - 0.470score
save-plan
danielrosehill
Save a Claude Code plan (from ExitPlanMode, design discussion, or multi-step task) to persistent storage.
SkillAFreshness: A★ 2Claude Code +9 - 0.470score
onboard
danielrosehill
First-run onboarding for the adb-ops plugin.
SkillAFreshness: A★ 2Claude Code +9 - 0.470score
context-gate
danielrosehill
One-shot context-checkpoint workflow — when the current session is getting cluttered, write a structured handover plan capturing what's done and what's…
SkillAFreshness: A★ 2Claude Code +9 - 0.470score
context-workflow-reference
danielrosehill
Reference and worked example for the CONTEXT.md / CLAUDE.md two-file workflow.
SkillAFreshness: A★ 2Claude Code +9 - 0.470score
scaffold-planning
danielrosehill
Create a lightweight `planning/` folder in the current repo with leftover-tasks.md, blockers.md, session-logs/, and handovers/.
SkillAFreshness: A★ 2Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get onboard running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
onboard skill. Agent Mart. https://agentmart.dev/skills/danielrosehill-onboard (last updated 2026-10-24)