Search / io.github.mrpkk/drift
protogridR0· no sign-inBlock an action a signed mandate does not permit: amount, payee, audience, currency, window, intent.
v0.1.0 · active · repository · website · descriptor JSON · versions
{
"mcpServers": {
"drift": {
"type": "http",
"url": "https://drift.rta.lat/mcp"
}
}
}claude mcp add --transport http drift https://drift.rta.lat/mcp[mcp_servers.drift]
url = "https://drift.rta.lat/mcp"
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"drift": {
"type": "remote",
"url": "https://drift.rta.lat/mcp",
"enabled": true
}
}
}{
"mcpServers": {
"drift": {
"type": "http",
"url": "https://drift.rta.lat/mcp"
}
},
"deeplink": "cursor://anysphere.cursor-deeplink/mcp/install?name=drift&config=eyJ0eXBlIjoiaHR0cCIsInVybCI6Imh0dHBzOi8vZHJpZnQucnRhLmxhdC9tY3AifQ%3D%3D"
}{
"servers": {
"drift": {
"type": "http",
"url": "https://drift.rta.lat/mcp"
}
}
}{
"mcpServers": {
"drift": {
"httpUrl": "https://drift.rta.lat/mcp"
}
}
}extensions:
"drift":
enabled: true
name: "drift"
type: streamable_http
uri: "https://drift.rta.lat/mcp"
timeout: 300
protocol.modernnewest supported version is 2025-06-18; 2026-07-28 not supported, older versions are deprecated until 2027-07-28protocol.statelessonly applies to 2026-07-28 serversprotocol.transportstreamable HTTPprotocol.list_ttlonly applies to 2026-07-28 serversauth.prm, auth.as_metadata, auth.cimdno remote uses OAuthauth.secret_in_urlno templated URLtools.descriptionsevery tool has a descriptiontools.description_length1 tools have descriptions over 1,024 characters, which crowds the contexttools.schemasevery tool has an object input schematools.annotations3 of 3 tools declare readOnlyHint or destructiveHinttools.directory_hintsevery tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHinttools.token_costabout 1,146 tokens to load every tooltools.api_dumptools are not a one-to-one API dumpstability.changes, stability.rug_pulltool history recorded for 3 of the 7 days neededdeps.known_vulns, deps.mcp_sdk_version, deps.resolvableno npm or PyPI packageNo tool definition changes recorded. A server's first observation is its baseline; later probes record what changes.
Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed. How it is computed.
+repository +namespace-matches-repo +website +reachable +uptime-83% +updated-4d-ago
Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed. How it is computed.
| remote | auth | reachable | uptime 30d | p50 | protocol | last ok |
|---|---|---|---|---|---|---|
https://drift.rta.lat/mcp streamable-http | none | yes | 83% | 1350 ms | 2025-06-18 | 2026-10-19 |
drift_checkread-only | Check ONE proposed action against a signed agent mandate BEFORE it happens. Returns {"verdict": "ALLOW"|"DENY", "rule": <name>} plus a report with a ledger fingerprint. Six rules run in a fixed order, so a refusal names the most important reason rather than an incidental one: amount, payee, audience, currency, window, intent. Anything unclassified is DENY. Use immediately before any action that spends money, moves a credential or changes state on someone else's behalf. Do NOT use to: verify mandate signatures (AP2 does that, and this server runs after signature verification), execute the action (it only decides), or certify that an intent is truthful — "intent" arrives from the agent and is untrusted input. An agent that lies within its limits still leaves a ledger entry; proving what really happened is attest's job, not this one's. Amounts are integer minor units (cents), never floats. Window bounds are inclusive. Fully idempotent: the same inputs always give the same verdict and the mandate is never mutated. |
drift_explainread-only | Return the six enforcement rules in the exact order they are applied, each with a plain-language reason it exists. Takes no mandate and no action. Use when a refusal did not make sense, or before explaining a denial to someone whose mandate was rejected. A refusal you cannot explain is one nobody will trust. Do NOT use it to check an action — it performs no check, and cannot predict which rule will fire, only the order they are tried in. Constant output, fully idempotent, opens no files, contacts no host. |
drift_sessionread-only | Check a SEQUENCE of proposed actions against one mandate. Returns {submitted, allowed, denied, breach_by_rule, fingerprint}. Use for a whole plan or batch, not a single action — use drift_check for that. The fingerprint is the point: rewriting or reordering any earlier decision changes the hash, so tampering with history becomes visible instead of silent. Do NOT use it to execute the actions, or as evidence of what actually happened. It reports what was PROPOSED against a mandate you supply. Read-only. Idempotent for a fixed action list. Amounts in integer minor units, as in drift_check. |
Schemas: list_tools.
{
"name": "io.github.mrpkk/drift",
"title": "drift — mandate enforcement for AI agents",
"$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
"remotes": [
{
"url": "https://drift.rta.lat/mcp",
"type": "streamable-http"
}
],
"version": "0.1.0",
"repository": {
"url": "https://github.com/mrpkk/drift",
"source": "github"
},
"websiteUrl": "https://drift.rta.lat/",
"description": "Block an action a signed mandate does not permit: amount, payee, audience, currency, window, intent."
}