AgentMesh Governance

AI agent governance, trust scoring, and policy enforcement powered by AgentMesh. Activate when: (1) user wants to enforce token limits, tool restrictions, or...

Security Scan

The skill's files and instructions generally match the stated governance purpose, but there are a few implementation and supply-chain risks (unsafe shell→python string interpolation, ambiguous key handling, and optional pip installs from external sources) that merit caution before installing or running.

Purpose & Capability

Name, description, and shipped scripts align with governance, trust scoring, identity verification, and audit logging. Requiring python3/pip is reasonable for the provided Python-based scripts. No unrelated env vars, binaries, or config paths are requested.

Instruction Scope

Runtime instructions are scoped to running the included scripts and (optionally) a backing 'agentmesh' Python package. However, the scripts embed user-supplied shell variables directly into python -c string literals (e.g., '$ACTION', '$DID', '$MESSAGE'), which can break or allow code injection if arguments contain unescaped quotes or malicious payloads. Also, generate-identity.sh warns to 'Store your private key securely' but the script does not expose or export the private key — this is confusing and could lead users to believe a private key was produced and saved when it was not.

Install Mechanism

No install spec is included (instruction-only), which is low risk. SKILL.md suggests 'pip install agentmesh' or installing from a GitHub repo via pip (git+https://github.com/imran-siddique/agent-mesh.git). Installing an external Python package from PyPI or a git repo will execute third-party code on the host — a normal pattern but a supply-chain risk that users should audit first. The GitHub URL itself is a common host, not an obscure or shortened URL.

Credentials

The skill requests no credentials or special config paths. The operations performed (reading a local policy.yaml, returning JSON results) are proportionate to the governance use-case. There are no hidden environment variable accesses in the scripts.

Persistence & Privilege

The skill does not request permanent/always presence and does not modify other skills or global agent config. It runs as on-demand scripts and optionally delegates to an external Python package for persistent functionality; this is consistent with the described behavior.

What to consider before installing

This skill appears to implement the claimed governance functions and the included scripts match the description. Before installing or running it: (1) review the upstream 'agentmesh' package source if you plan to pip install it — installing packages runs remote code; (2) avoid passing untrusted or unsanitized input to the scripts because they interpolate shell args directly into python -c string literals (quotes or special characters could break the Python code or be exploited); (3) note that generate-identity.sh does not emit or save a private key despite its warning — verify key handling if you need persistent key material; (4) run the scripts in a sandboxed environment or container the first time, and audit the referenced GitHub repo and package code if you intend to rely on persistent trust/identity functionality.

AgentMesh Governance — Trust & Policy for OpenClaw Agents

Zero-trust governance layer for OpenClaw agents. Enforce policies, verify identities, score trust, and maintain tamper-evident audit logs — all from your agent's command line.

Setup

Install the AgentMesh governance CLI:

pip install agentmesh-governance

If agentmesh-governance is not yet on PyPI, install directly from source:

pip install "agentmesh @ git+https://github.com/imran-siddique/agent-mesh.git"

Scripts

All scripts are in scripts/. They wrap the governance engine and output JSON results.

Check Policy Compliance

Evaluate an action against a governance policy before execution:

scripts/check-policy.sh --action "web_search" --tokens 1500 --policy policy.yaml

Returns JSON with allowed: true/false, any violations, and recommendations. Use this before executing any tool call to enforce limits.

Get Trust Score

Check an agent's current trust score (0.0 – 1.0):

scripts/trust-score.sh --agent "research-agent"

Returns the composite trust score with breakdown across 5 dimensions: policy compliance, resource efficiency, output quality, security posture, collaboration health.

Verify Agent Identity

Verify an agent's Ed25519 cryptographic identity before trusting its output:

scripts/verify-identity.sh --did "did:agentmesh:abc123" --message "hello" --signature "base64sig"

Returns verified: true/false. Use when receiving data from another agent.

Record Interaction

Update trust scores after collaborating with another agent:

scripts/record-interaction.sh --agent "writer-agent" --outcome success
scripts/record-interaction.sh --agent "writer-agent" --outcome failure --severity 0.1

Success adds +0.01 to trust score. Failure subtracts the severity value. Agents dropping below the minimum threshold (default 0.5) are auto-blocked.

Audit Log

View tamper-evident audit trail with Merkle chain verification:

scripts/audit-log.sh --last 20
scripts/audit-log.sh --agent "research-agent" --verify

The --verify flag checks Merkle chain integrity — any tampering is detected.

Generate Identity

Create a new Ed25519 cryptographic identity (DID) for your agent:

scripts/generate-identity.sh --name "my-agent" --capabilities "search,summarize,write"

Returns your agent's DID, public key, and capability manifest.

Policy File Format

Create a policy.yaml to define governance rules:

name: production-policy
max_tokens: 4096
max_tool_calls: 10
allowed_tools:
  - web_search
  - file_read
  - summarize
blocked_tools:
  - shell_exec
  - file_delete
blocked_patterns:
  - "rm -rf"
  - "DROP TABLE"
  - "BEGIN CERTIFICATE"
confidence_threshold: 0.7
require_human_approval: false

When to Use This Skill

• Before tool execution: Run check-policy.sh to enforce limits
• Before trusting another agent's output: Run verify-identity.sh
• After collaboration: Run record-interaction.sh to update trust
• Before delegation: Check trust-score.sh — don't delegate to agents below 0.5
• For compliance: Run audit-log.sh --verify to prove execution integrity
• On setup: Run generate-identity.sh to create your agent's DID

What It Enforces

Policy	Description
Token limits	Cap per-action and per-session token usage
Tool allowlists	Only explicitly permitted tools can execute
Tool blocklists	Dangerous tools are blocked regardless
Content patterns	Block regex patterns (secrets, destructive commands, PII)
Trust thresholds	Minimum trust score required for delegation
Human approval	Gate critical actions behind human confirmation

Architecture

This skill bridges the OpenClaw agent runtime with the AgentMesh governance engine:

OpenClaw Agent → SKILL.md scripts → AgentMesh Engine
                                     ├── GovernancePolicy (enforcement)
                                     ├── TrustEngine (5-dimension scoring)
                                     ├── AgentIdentity (Ed25519 DIDs)
                                     └── MerkleAuditChain (tamper-evident logs)

Part of the Agent Ecosystem: AgentMesh · Agent OS · Agent SRE