AffixIO Agent Tools (Model repo)

Fail-closed local tool layer: identity + permission checks, transaction intent validation, email search/organisation, drafting and sending with human approval, webhook/API actions behind allow-lists, and an audit record for every decision. No network, no production keys.

from tools import run_tool, approve
print(run_tool("email.search", {"query": "invoice"}))
print(run_tool("email.send", {"to": "boss@acme.com", "subject": "hi", "body": "ok"}))
print(approve("<appr_id>", True))

Tools: identity.check, permissions.check, payments.validate, email.search, email.organise, email.draft, email.send (approval), webhook.trigger (approval + allow-list), api.call (approval + allow-list), audit.show. Tests: python test_tools.py.

Version

  • v1.0.0 — initial release. See README.md.

Production path (paid)

Simulated receipts, not cryptographic attestations. Hosted verification, ML-DSA attestations, team agents/inboxes, enterprise audit:

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Collection including AffixIO/affixio-agent-tools