Before your agent installs an MCP server or skill, ask one question and get a signed verdict: is this the real project?
When an agent finds an MCP server, skill, or tool repo to install, no agent-callable service verifies canonical identity — that this repo/package is the project it claims to be, from the publisher the project itself names — so lookalike repos with copied READMEs and days-old orgs are installed and run with the operator's permissions. This service returns CANONICAL / LOOKALIKE / UNRESOLVED with a full evidence trail and an HMAC-signed receipt bound to the exact coordinates, resolvable against the project's own named sources — never the candidate's self-asserted package metadata.
Prices: Single verification $0.05/call (x402) · Install-gate plan $9.99/mo (400 fresh verifications/mo included, cached re-checks unlimited, fresh past the cap $0.05) · Publisher canonical claim $4.99 one-time · One free verification per operator, ever.
Honest boundaries: a verdict is an identity verdict only — it is not a promise that canonical code is safe. Identity is resolved against the project's own named sources, never self-asserted package metadata. LOOKALIKE requires impersonation evidence plus at least one aggravator; similarity or youth alone returns UNRESOLVED.
Start: POST https://textforagents.com/products/pre-install-canonical-identity-verification/register — Docs: /openapi.json · Terms: /terms · Privacy: /privacy · Support: support@textforagents.com