---

name: pin-setup description: Use when initializing the pins knowledge graph for the first time in a project. Guides the user through creating pins.yaml (the storage manifest). Triggers on "setup pins", "initialize pins", "create pins.yaml", "first-run pins".

pin-setup

First-run interview to create ~/.pins/{name}/pins.yaml, the storage manifest for the pins knowledge graph.

What is pins.yaml

pins.yaml is the manifest that tells the pins API where to read and write pin files. It declares the storage location and scope for this project's knowledge graph.

resolveManifest() (from lib/pins/manifest.ts) reads pins.yaml and returns the parsed manifest. Callers then extract manifest.location and pass it explicitly to each API (record, query, audit, etc.). The manifest itself is advisory metadata — individual API calls always receive location as an explicit argument, not by re-reading the file.

Interview

Walk the user through three decisions:

1. Storage location

The manifest is always written to ~/.pins/{name}/pins.yaml, where {name} is derived from the project name of the cwd (deriveProjectName(cwd)). This path is fixed.

Ask: "Where should pin data files live?"

Common choices:

• Co-located with the manifest (default) — ~/.pins/{name}/ — no --location needed
• A custom absolute path — pass via --location; useful for monorepos or multi-project setups

Confirm whether the directory should be created if it does not exist (it will be, automatically, on the first record() call).

2. Scope

Ask: "What is the scope of this pins manifest — personal (private) or project-wide (shared)?"

• private — pins default to sensitivity: private; not intended for team consumption
• shared — pins default to sensitivity: shared; meant to be read by teammates

The scope sets the manifest-level default; individual pins can override sensitivity in their frontmatter.

3. Git management

Ask: "Is this pins corpus managed under git?"

This is independent of storage location: a ~/.omt/... directory can be a git repo; a .pins/ directory inside a project repo can be gitignored. The answer reflects intent, not directory placement. When git: true, the SessionStart hook and the pin-record/pin-wrap-up skills will prompt the AI to commit pin file changes after recording.

Record the answer as git: true or git: false.

Output

Once all three questions are answered, write pins.yaml to ~/.pins/{name}/pins.yaml:

# pins.yaml — knowledge graph storage manifest
location: <resolved absolute path>
scope: <private|shared>
git: <true|false>

Confirm the path is correct and the user understands that all pins API calls will resolve against this manifest.

Migration

If the project has legacy pin files (frontmatter with slug instead of id, no type field), run migration after setup:

bun "${CLAUDE_SKILL_DIR}/scripts/setup.ts" --scope "$SCOPE" [--location "$LOC"] [--git true|false]

--location is optional. When omitted, pin data is co-located with the manifest at ~/.pins/{name}/. The script writes pins.yaml to ~/.pins/{name}/pins.yaml and then runs migrate() against the just-written location. migrate() is idempotent — re-running it is safe. Each legacy file gets a .bak sibling before conversion.