/foundry:index — the machinery catalog

A thin, read-only index of every piece of foundry machinery — each skill, playbook, and agent with its trigger — so the operator can see what exists and how to invoke it. It is the operational face of the Terminology Standard: verb → machinery → example prompt.

It derives the catalog from the skills/*/SKILL.md + agents/*.md frontmatter at run time (the source of truth), so it can never go stale — there is no committed catalog to maintain. It is not a driver: it lists; it never runs anything.

When to trigger

• "what can I run?", "list the foundry machinery / skills / commands", "show the catalog / index", "what playbooks are there?", "/foundry:index".

Use

python3 "${CLAUDE_PLUGIN_ROOT}/scripts/foundry-index.py"          # rendered markdown catalog
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/foundry-index.py" --json   # the data form

Output is a name | machinery | trigger table (machinery ∈ skill / playbook / agent) plus a one-line count summary. The trigger column is the first segment of each item's description — where, per the standard, the canonical verb-object and natural-language trigger phrases live. To trigger an item, use its verb-object (/foundry:<name>) or one of its natural-language phrases.

What it does NOT do

• It never invokes anything — it is a read-only catalog, not a runner.
• It maintains no stored list — the catalog is computed from frontmatter on every run, so it cannot drift from the actual machinery.