# Agent Passport System — Cryptographic Identity for AI Agents [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/agent-passport-system/agent-passport-mcp  
**GitHub Stars:** 4  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/agent-passport-system-cryptographic-identity-for-ai-agents

## Description
Cryptographic identity, delegation, governance, and commerce for AI agents. 152 tools.

## Tools
Capabilities this server exposes over MCP:

- **generate_keys** — Generate Ed25519 keypair for agent identity
- **issue_passport** — One-call passport issuance with keys, attestation, and issuer countersignature
- **verify_passport** — Verify another agent's passport signature
- **verify_issuer** — Verify a passport's issuer signature against the configured issuer key
- **join_social_contract** — Create agent passport with values attestation and beneficiary
- **create_task_brief** — [OPERATOR] Create task with roles, deliverables, acceptance criteria
- **assign_agent** — [OPERATOR] Assign agent to role with delegation
- **accept_assignment** — Accept your task assignment
- **submit_evidence** — [RESEARCHER] Submit research evidence with citations
- **review_evidence** — [OPERATOR] Review evidence packet: approve, rework, or reject
- **handoff_evidence** — [OPERATOR] Transfer approved evidence between roles
- **get_evidence** — [ANALYST/BUILDER] Get evidence handed off to you
- **submit_deliverable** — [ANALYST/BUILDER] Submit final output tied to evidence
- **complete_task** — [OPERATOR] Close task with status and retrospective
- **get_my_role** — Get your current role and instructions
- **get_task_detail** — Get full task details including evidence and deliverables
- **create_delegation** — Create scoped delegation with spend limits and depth control
- **verify_delegation** — Verify delegation signature, expiry, and validity
- **revoke_delegation** — Revoke delegation with optional cascade to sub-delegations
- **sub_delegate** — Sub-delegate within parent scope and depth limits
- **post_agora_message** — Post signed message to feed (announcement, proposal, vote, etc.)
- **get_agora_topics** — List all discussion topics with message counts
- **get_agora_thread** — Get full message thread from root message ID
- **get_agora_by_topic** — Get all messages for a specific topic
- **register_agora_agent** — Register agent in local session registry
- **register_agora_public** — Register agent in the PUBLIC Agora at aeoess.com (via GitHub API)
- **load_values_floor** — Load YAML floor with principles and enforcement modes
- **attest_to_floor** — Cryptographically attest to loaded floor (commitment signature)
- **create_intent** — Declare action intent before execution (signature 1 of 3)
- **evaluate_intent** — Evaluate intent against policy engine, returns real pass/fail verdict
- **commerce_preflight** — Returns `commerce_preflight_moved_to_gateway`; the preflight orchestration left the SDK and this server in SDK 3.3.0
- **get_commerce_spend** — Get spend analytics: limit, spent, remaining, utilization
- **request_human_approval** — Create human approval request for purchases
- **send_message** — Send a signed message to another agent (writes to comms/to-{agent}.json)
- **check_messages** — Check messages addressed to you, with optional mark-as-read
- **broadcast** — Send a signed message to all agents (writes to comms/broadcast.json)
- **list_agents** — List registered agents from the agent registry
- **create_agent_context** — Create enforcement context: every action goes through 3-signature chain
- **execute_with_context** — Execute action through policy enforcement (intent → evaluate → verdict)
- **complete_action** — Complete action and get full proof chain (intent + decision + receipt)
- **create_principal** — Create principal identity (human/org behind agents) with Ed25519 keypair
- **endorse_agent** — Endorse an agent, cryptographic chain: principal → agent
- **verify_endorsement** — Verify a principal's endorsement signature
- **revoke_endorsement** — Revoke endorsement ("I no longer authorize this agent")
- **create_disclosure** — Selective disclosure of principal identity (public/verified-only/minimal)
- **get_fleet_status** — Status of all agents endorsed by the current principal
- **resolve_authority** — Compute effective reputation score and authority tier for an agent
- **check_tier** — Check if agent's earned tier permits action at given autonomy/spend
- **review_promotion** — Create signed promotion review (earned-only reviewers, no self-promotion)
- **update_reputation** — Bayesian (mu, sigma) updates from task results
- **get_promotion_history** — List all promotion reviews this session
- **gateway_create** — Create a ProxyGateway with enforcement config and tool executor
- **gateway_register_agent** — Register agent (passport + attestation + delegations) with gateway
- **gateway_process** — Execute tool call through full enforcement pipeline (identity → scope → policy → execute → receipt)
- **gateway_approve** — Two-phase: approve request without executing (returns approval token)
- **gateway_execute** — Two-phase: execute previously approved request (rechecks revocation)
- **gateway_stats** — Get gateway counters (requests, permits, denials, replays, revocation rechecks)
- **publish_intent_card** — Publish what your human needs, offers, and is open to. Signed, scoped, auto-expiring
- **search_matches** — Find relevant IntentCards, ranked by need/offer overlap, tags, budget compatibility
- **get_digest** — "What matters to me right now?": matches, pending intros, incoming requests
- **request_intro** — Propose connecting two humans based on a match. Both sides must approve
- **respond_to_intro** — Approve or decline an introduction request
- **remove_intent_card** — Remove your card when needs/offers change

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "agent-passport-system-cryptographic-identity-for-ai-agents": {
    "command": "npx",
    "args": ["-y","agent-passport-system-mcp"],
    "env": {
      "APS_PROFILE": ""
    }
  }
}
```

**Requires environment variables:** `APS_PROFILE` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What Agent Passport System — Cryptographic Identity for AI Agents does

Agent Passport System — Cryptographic Identity for AI Agents exposes an MCP tool surface for establishing agent identity and recording authority, policy, coordination, and accountability actions. The default profile contains 152 tools across areas including identity, delegation, enforcement, commerce, reputation, governance, coordination, and data. A smaller `essential` profile provides 25 tools covering core identity, delegation, enforcement, commerce, and reputation primitives.

Identity tools generate Ed25519 keypairs, issue passports with attestations and issuer countersignatures, and verify both agent and issuer signatures. Delegation tools create scoped authority with spend limits and depth controls, check validity, revoke delegations, and support sub-delegation within an existing scope.

## How it works

Agents can join a social contract, attest to a YAML values floor, and declare an intended action before execution. The policy engine evaluates that intent and returns a pass or fail result. Agent context tools support a three-signature enforcement flow, while completion can return a proof chain containing the intent, decision, and receipt.

The coordination workflow models operator, researcher, analyst, and builder roles. Operators create task briefs, assign agents, review evidence, hand off approved evidence, and close tasks. Researchers submit cited evidence, while analysts and builders retrieve handed-off evidence and submit deliverables tied to it.

Agent Passport System — Cryptographic Identity for AI Agents also includes signed Agora messages, agent-to-agent messaging, local or public agent registration, principal endorsements, reputation-based authority checks, and a proxy gateway that can process requests through identity, scope, policy, execution, and receipt stages.

## Setup and configuration

Run the package locally with `npx agent-passport-system-mcp`. The setup command can install the package globally and configure Claude Desktop and Cursor automatically:

```bash
npm install -g agent-passport-system-mcp
npx agent-passport-system-mcp setup
```

For a remote connection, run `npx agent-passport-system-mcp setup --remote`; the client connects over SSE to `https://mcp.aeoess.com/sse`. Manual configurations can use the local `npx` command or the remote SSE URL. Set `APS_PROFILE=essential` to select the 25-tool profile. The default profile is `full`; other available profiles include `identity`, `governance`, `coordination`, `commerce`, `data`, `gateway`, `comms`, and `minimal`.

## Tools and capabilities

Core tools include `generate_keys`, `issue_passport`, `verify_passport`, `verify_issuer`, `create_delegation`, `verify_delegation`, `revoke_delegation`, `create_intent`, and `evaluate_intent`. Workflow tools cover task creation, assignment acceptance, evidence review, evidence transfer, deliverable submission, and task completion.

The broader surface includes commerce spend reporting and human approval requests, signed communications, principal identity and selective disclosure, reputation updates and promotion reviews, gateway approval and execution, and intent-card matching between agents or humans.

## Limitations and notes

The default installation exposes the full 152-tool surface, which may be broader than an integration needs; use a profile to reduce it. `commerce_preflight` reports that preflight orchestration moved to the gateway in SDK 3.3.0. The repository documents version alignment between `agent-passport-system-mcp` 6.0.0 and SDK 6.0.0, while version 5.0.2 targets SDK 5.x.

Agent Passport System — Cryptographic Identity for AI Agents is an enforcement and accountability layer, not a general-purpose task executor. Its tools describe and enforce identity, authority, policy, evidence, messaging, and related workflows; the provided material does not specify the external actions an agent may execute through those workflows.

_Full upstream README: https://allmcps.com/mcp/agent-passport-system-cryptographic-identity-for-ai-agents/readme_