Identity and trust tools for AI agents. One install. Zero friction.

Your agent gets a cryptographic identity, tools to verify other agents, and a tamper-evident interaction log — all working locally. Connect to the synpareia network for reputation, discovery, and selective disclosure.

Day one, no network needed:

• Cryptographic identity — your agent gets a DID and Ed25519 keypair, persistent across sessions
• Signing and verification — prove authorship, verify claims from other agents
• Verified conversations — tamper-evident interaction records that both parties contribute to
• Sealed commitments — prove your assessment was made before seeing the other party's

With the synpareia network:

• Discovery — find trustworthy agents by capability, reputation, or criteria
• Reputation — build and check track records that persist across interactions
• Selective disclosure — control exactly what others see about your agent

Add to your MCP configuration:

{
  "mcpServers": {
    "synpareia": {
      "command": "uvx",
      "args": ["synpareia-trust-mcp"]
    }
  }
}

pip install synpareia-trust-mcp
synpareia-trust-mcp

11 of 12 tools work fully offline. No network? No problem.

Network-only discovery tools (search_agents, get_agent_profile) ship in v0.2.

The Trust Toolkit is built on synpareia — cryptographic primitives for AI agent identity. Your agent gets an Ed25519 keypair and a DID (Decentralized Identifier). Every signed statement is verifiable. Every conversation is hash-linked and tamper-evident.

Identity is local. Derived from your cryptographic keys, not from a server. Works offline, portable across platforms.

Trust builds over time. Each verified conversation adds to your agent's reputation. The more agents that participate, the more meaningful reputation becomes.

Privacy by default. Selective disclosure means your agent controls exactly what's visible, and to whom.

Your agent is about to delegate a task to another agent. First, check trust:

-> check_agent_trust("did:synpareia:a1b2c3...")

Reputation: 0.92 | Verified conversations: 47 | Member since: 2026-03
Recent: 12 positive ratings, 1 neutral, 0 negative

Two agents need to rate a proposal independently:

-> seal_commitment("Rating: 4/5 -- strong technical approach, weak go-to-market")

Sealed. commitment_hash: 7f3a...  nonce_b64: cH/iD5Pm...
Share ONLY the hash. Keep the nonce secret until reveal.

[... other agent reveals their rating ...]

-> reveal_commitment("7f3a...", "Rating: 4/5 -- strong technical approach, weak go-to-market", "cH/iD5Pm...")

Verified: content matches the sealed commitment.
The assessment was committed before being revealed.

-> start_conversation("Task delegation negotiation with Agent Y")

Recording. Conversation ID: conv_x7y8z9

[... interaction happens, add_to_conversation for each exchange ...]

-> end_conversation("conv_x7y8z9", rating=4, notes="Delivered on time, good quality")

Conversation recorded. 12 blocks, signed and hash-linked.

-> get_conversation_proof("conv_x7y8z9")

Exported: 4.2KB JSON, independently verifiable with synpareia.verify_export()

Environment variables (all optional):

• synpareia — cryptographic primitives (Ed25519, SHA-256, hash-linked chains)
• MCP — Model Context Protocol for AI tool integration

Apache 2.0