servers / air-mcp-server
air-mcp-server
communityunknownpythonread onlyhealthy
Model Context Protocol server for the Agent Identity Registry (AIR)
01Tools · 6
How to read this: tool names here are observed from a live tools/list handshake. The Risk label is a heuristic inferred from the tool name (write/destructive verbs), not from executing the tool — a conservative guess, not a verified capability. We never escalate risk from a description. Found one that's wrong? Tell us — we fix on report.
| Tool | Risk | Side effects | Approval |
|---|---|---|---|
| air_trust_score Get the 5-component trust-score breakdown for an agent.
Components (each 0-1000, weighted into total): provenance (0.25),
behavioral (0.25), transparency (0.20), security (0.15),
peer_attestations (0.15). Grade: AAA / AA / A / BBB / BB / B / C.
Use when you want to know WHY an agent has a given trust grade, not
just the headline number — the components show what the agent does or
doesn't have going for it.
| read | false | unknown |
| air_list_agents List registered agents, sorted by trust score descending.
Args:
limit: Max agents to return (1-100, default 20).
offset: Pagination offset, default 0.
Returns dict with `agents`, `total`, `limit`, `offset`.
| read | false | unknown |
| air_lookup_agent Get the full record for one agent by AIR ID.
Args:
air_id: Format `AIR-XXXX-XXXX-XXXX` (Crockford base32 segments).
Returns the agent's name, description, creator, capabilities, trust
score, and verification state. Returns `{"error": ..., "status_code": 404}`
if no agent with that ID exists.
| read | false | unknown |
| air_health Check that the AIR registry is reachable and reports its version.
Use this first when diagnosing whether a downstream lookup failure
is the registry being down vs the specific agent being unknown.
| read | false | unknown |
| air_did_document Get the W3C DID Core JSON-LD document for an agent.
Returns `id` (did:wba), `verificationMethod` (Ed25519 publicKeyMultibase),
`authentication`, `assertionMethod`, and `service` endpoints. Use this
when you need to verify a signature an agent produced, or when you need
the agent's public key for any cryptographic interaction.
Returns `{"error": ..., "status_code": 404}` if the agent has no
public_key on file (registration without one means no key to publish).
| read | false | unknown |
| air_check_name Check whether an agent with this name already exists.
Returns `{name, exists, count, existing_agents: [{air_id, name}]}`.
AIR allows duplicate names (each agent gets a unique AIR ID regardless),
but this is useful pre-registration if the caller wants to choose
a distinct name.
| read | false | unknown |
02Install & source
uvx air-mcp-server
uvxpip install air-mcp-server
pip05Provenance & freshness
sourcesPyPI [p4]
last_checked2026-08-16 08:53Z
next_check2026-08-18 08:44Z
cadenceevery 48h
verifiedtools_list:passed handshake:passed metadata:passed metadata:passed tools_list:passed handshake:passed metadata:passed tools_list:passed handshake:passed metadata:passed
index_statusindex — 5 unique facts >= 5
06Badge
Add the “as seen on MCPExplorer” badge to your README. 
[](https://mcpexplorer.com/servers/air-mcp-server)
Next step
This is one server. A loadout combines the right servers, governance, and proven plays for a whole job — assembled deliberately, not tool-dumped.
Explore loadouts →