تخطَّ إلى المحتوى
TakoAPI
العودة إلى المهارات

agentmesh-governance

بواسطة imran-siddique

Coding Agents & IDEs
495التنزيلات وفق ما يذكره clawskills.sh
0المشاهدات
لا إعجاباتسجّل الدخول للإعجاب بهذه المهارة
—لم يُقيَّم بعد
تقييمك

موجز

AI agent governance, trust scoring, and policy enforcement powered by AgentMesh.

الوصف

ما الذي تفعله هذه المهارة

AgentMesh Governance adds a zero-trust layer to OpenClaw agents, enforcing token limits, tool allowlists/blocklists, and content policies before execution. Agents get Ed25519 cryptographic identities (DIDs) and trust scores across five dimensions. All actions are recorded in tamper-evident Merkle chain audit logs.

Unlike ad-hoc prompt-based restrictions, this enforces governance cryptographically and tracks trust scores over time, making it auditable and tamper-evident rather than easily bypassed.

متى تستخدمها

  • Blocking a research agent from running shell commands via policy file
  • Verifying another agent's DID before accepting its summarized output
  • Checking trust score before delegating a subtask to a writer agent
  • Proving execution integrity to a compliance team with Merkle-verified audit logs
  • Auto-blocking agents that repeatedly fail collaborative interactions

مثال على سير العمل

الإدخال

User asks: enforce policy on a web_search action with 1500 tokens before executing

الوكيل
  1. 1

    Run check-policy.sh with --action web_search --tokens 1500 --policy policy.yaml

  2. 2

    Governance engine evaluates action against max_tokens, allowed_tools, and blocked_patterns

  3. 3

    Returns JSON with allowed: true/false and any violations

  4. 4

    If allowed, proceed with tool execution

  5. 5

    Run record-interaction.sh after completion to update trust score

الإخراج

JSON response: {"allowed": true, "violations": [], "recommendations": []}

التثبيت

clawhub install imran-siddique/agentmesh-governance

شغّل أمر واجهة سطر الأوامر، أو الصق النص في Claude Code أو Cursor أو Codex أو أي وكيل برمجة تستخدمه.

الروابط