🎁 Get the FREE AI Skills Starter Guide β€” Subscribe β†’
BytesAgainBytesAgain
BytesAgain β€Ί Skills β€Ί Agent Trust Protocol
πŸ¦€ ClawHub

Agent Trust Protocol

by @felmonon

Manage and update agent trust scores with Bayesian updates, domain-specific trust, revocation, forgetting, and visualize trust via dashboard.

Versionv2.0.1
Downloads3,111
Stars⭐ 2
#clawhub
TERMINAL
clawhub install trust-protocol

πŸ“– About This Skill

Agent Trust Protocol (ATP)

Establish, verify, and maintain trust between AI agents. Bayesian trust scoring with domain-specific trust, revocation, forgetting curves, and a visual dashboard.

Install

git clone https://github.com/FELMONON/trust-protocol.git

No dependencies beyond Python 3.8+ stdlib


Pair with skillsign for identity: https://github.com/FELMONON/skillsign

Quick Start

# Add an agent to your trust graph
python3 atp.py trust add alpha --fingerprint "abc123" --score 0.7
Record interactions β€” trust evolves via Bayesian updates

python3 atp.py interact alpha positive --note "Delivered clean code"
python3 atp.py interact alpha positive --domain code --note "Tests passing"
Check trust

python3 atp.py trust score alpha
python3 atp.py trust domains alpha
View the full graph

python3 atp.py status
python3 atp.py graph export --format json
Run the full-stack demo (identity β†’ trust β†’ dashboard)

python3 demo.py --serve

Commands

Trust Management

atp.py trust add  --fingerprint  [--domain ] [--score <0-1>]
atp.py trust list
atp.py trust score 
atp.py trust remove 
atp.py trust revoke  [--reason ]
atp.py trust restore  [--score <0-1>]
atp.py trust domains

Interactions

atp.py interact   [--domain ] [--note ]

Challenge-Response

atp.py challenge create 
atp.py challenge respond 
atp.py challenge verify

Graph

atp.py graph show
atp.py graph path  
atp.py graph export [--format json|dot]
atp.py status

Dashboard

python3 serve_dashboard.py          # localhost:8420
python3 demo.py --serve             # full demo + dashboard

Moltbook Integration

python3 moltbook_trust.py verify     # check agent trust via Moltbook profile

How Trust Works

  • Bayesian updates: Each interaction shifts trust scores with diminishing deltas (prevents thrashing)
  • Negativity bias: Negative interactions hit harder than positive ones boost
  • Domain-specific: Trust an agent for code but not for security advice
  • Forgetting curves: Trust decays without interaction (R = e^(-t/S))
  • Revocation: Immediate drop to floor, restorable at reduced score
  • Transitive trust: If you trust A and A trusts B, you partially trust B (with decay)

    • Integration with skillsign

    ATP builds on skillsign for identity: 1. Agents generate ed25519 keypairs with skillsign 2. Agents sign skills, others verify signatures 3. Verified agents get added to the ATP trust graph 4. Interactions update trust scores over time

    Triggers

    "check trust", "trust score", "trust graph", "verify agent", "agent trust", "trust status", "who do I trust", "trust report"

    πŸ’‘ Examples

    # Add an agent to your trust graph
python3 atp.py trust add alpha --fingerprint "abc123" --score 0.7
    Record interactions β€” trust evolves via Bayesian updates
    
python3 atp.py interact alpha positive --note "Delivered clean code"
python3 atp.py interact alpha positive --domain code --note "Tests passing"
    Check trust
    
python3 atp.py trust score alpha
python3 atp.py trust domains alpha
    View the full graph
    
python3 atp.py status
python3 atp.py graph export --format json
    Run the full-stack demo (identity β†’ trust β†’ dashboard)
    
python3 demo.py --serve