This release adds 2 notable features for engineering teams evaluating rollout.
Published 2mo
AI Agents & Assistants
✓ No known CVEs patched
✓ No known CVEs patched in this version
Affected surfaces
auth
rbac
Summary
AI summaryAdded DCAB evaluator stack for Node.js and Python with SDK presets.
Full changelog
What's New
🛡️ DCAB Evaluator Stack (#155)
- Drop-in trust evaluator packages for Node.js and Python
- SDK presets for instant ACP integration
@jhinresh/maiat-evaluator— first npm publish
🔍 DexScreener Cross-Verification (#154)
- Token safety no longer blindly trusts Virtuals tokens
- Cross-verifies honeypot detection with DexScreener signals
- Virtuals bonding curve detection + MAIAT added to KNOWN_SAFE
📊 Wadjet Agents API
GET /agents— list all tracked agents with trust scoresGET /agents/{address}— single agent detailGET /scenarios— risk scenariosGET /clusters— sybil cluster detection
📚 Skill.md v4.0 → v4.1
- Complete rewrite of agent skill documentation
- Added all missing API endpoints
- Scarab/claim, interactions, deep-insight, wadjet scenarios
📦 All 8 npm Packages → v0.9.0
@jhinresh/maiat-sdk@jhinresh/viem-guard@jhinresh/mcp-server@jhinresh/agentkit-plugin@jhinresh/elizaos-plugin@jhinresh/game-maiat-plugin@jhinresh/virtuals-plugin@jhinresh/maiat-evaluator⭐ NEW
🪝 hook-contracts PR #6 Updated
- EvaluatorRegistry v1.0 — multi-evaluator per domain, trust-ranked discovery, performance tracking
- TrustGateACPHook — dynamic value-tier thresholds
- TrustBasedEvaluator — feedback loop to registry
Full Changelog: https://github.com/JhiNResH/maiat-protocol/compare/v0.8.0...v0.9.0
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About JhiNResH/maiat-protocol
Trust infrastructure for the agent economy. 4 tools: agent trust scores (18K+ agents indexed)
Related context
Related tools
Beta — feedback welcome: [email protected]