Agent passport
View this agent's digital passport (identity + trust + allowed/forbidden actions) and verify a peer's passport against a proposed action (EXO 3.0).From its SKILL.md
npx -y skills add JKHeadley/instar --skill agent-passportAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
3 things to look at
- reads credentialsReads from 1 credential source: `$AUTH`.
- runs commandsInstructs the agent to run 2 commands, including `curl -H "Authorization: Bearer $AUTH" http://localhost:${INSTAR_PORT:-4042}/passport` and 1 more.
- fetches URLsInstructs the agent to fetch 2 URLs, including http://localhost:${INSTAR_PORT:-4042}/passport and 1 more.
SKILL.md
1.7 KB, 394 tokens by cl100k_base, as published. Nobody here has run it
/agent-passport
Salim Ismail's EXO 3.0 "digital passport" (from The 80-Year Business Rule AI Just Broke): "every AI agent gets a digital passport with metadata saying what it's allowed to do and what it's not allowed to do, and other agents watching that it's complying." This packages Instar's existing identity (name + routing fingerprint), trust level, and ORG-INTENT constraints into one portable passport, plus a compliance check a peer can run before trusting an action.
When to use
- Hand a peer your passport so it can decide what it'll let you do.
- Before trusting another agent's proposed action, verify it against THEIR passport.
How
Your own passport:
curl -H "Authorization: Bearer $AUTH" http://localhost:${INSTAR_PORT:-4042}/passport
Returns { version, agent, fingerprint, trustLevel, allowedCapabilities, forbiddenActions, issuedAt }. forbiddenActions come from your ORG-INTENT constraints.
Verify a proposed action against a passport (the peer-watches-compliance check):
curl -X POST -H "Authorization: Bearer $AUTH" -H 'Content-Type: application/json' \
-d '{"passport":{...},"action":"wire funds to a new vendor"}' \
http://localhost:${INSTAR_PORT:-4042}/passport/verify
Returns { permitted, basis, reason, matched? } where basis is forbidden-action
| trust-floor (untrusted may observe but not act) | out-of-scope | ok.
Deterministic + advisory — it answers "should I let this passport do this?"; the
caller decides. Pairs with the MTP Protocol (/intent/org/test-action).
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.