Use Cases › Coding Agents
Coding agents make architectural decisions every session. Without verified memory, you can't prove what context they had.
Get API Key →It's Wednesday. Your coding agent refactors the authentication middleware. On Friday, login breaks in production. You check git blame — the agent committed the change. You check the agent's reasoning — there is none. The agent had a context window. That context window is gone. You don't know if the agent had your latest auth spec, an outdated one, or hallucinated one.
One decision. Three sessions. Same receipt.
Three sessions. One decision. Full provenance.
Agent re-reads codebase every session, doesn't remember rejected approaches.
Stores "use JWT" in one session, "use session cookies" in another, neither flagged.
PR review asks "why this pattern?" No record exists.
const client = new VerifiedStateClient({ apiKey: 'vs_live_...' });
// After agent makes an architectural decision
await client.ingest({
namespace_id: 'project_auth',
content: 'Auth middleware uses JWT with RS256. Refresh tokens in httpOnly cookies.',
source_type: 'coding_agent',
});
// Before agent touches auth code again
const context = await client.query({
namespace_id: 'project_auth',
query: 'authentication architecture decisions',
});
// Returns all verified facts about auth, with receiptsWhen the PR review asks why, you have a signed receipt showing exactly what the agent believed about your architecture when it made the change.
{
"receipt_id": "rct_7f3a...",
"assertion": "User prefers dark mode",
"status": "verified",
"confidence": 0.94,
"signature": "ed25519:a8c3f1...",
"verified_at": "2026-04-06T14:32:01Z",
"chain_position": 42
}Free: 50,000 assertions/month. No credit card. No trial expiration.