- Midplane @midplane/midplaneSafe-by-default SQL guardrails for AI agents: AST-checked queries, per-table policy, audit log.2% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.5 GitHub stars.
- THOR @nworks3d/thorLocal memory for coding agents; a rule with a proof can refuse. Full install with hooks: the repo.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.5 GitHub stars.
- valv @valv/mcpLet coding agents query any SQL database safely. Read-only by default and scoped by your policies.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.5 GitHub stars.
- AgenticLedger @shekharbhardwaj/agentic-ledgerLocal-first flight recorder for AI agents: sessions, costs, loop runs, and stuck-loop flags.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.5 GitHub stars.
- Agent Abilities for MCP @unaibamir/agent-abilities-for-mcpSelf-hosted WordPress MCP server with per-capability permission controls and a full audit log.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.5 GitHub stars.
- Pacioli — the ERPNext broker you can hand the books @john-broadway/pacioliGoverned agent access to ERPNext — MCP + A2A doors, one spine. No debit without a credit.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.5 GitHub stars.
- Vigilis @piyushpathakqa/vigilis-mcpGate AI-written tests: heal safe drift, refuse to hide real bugs, sign a verifiable receipt.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.5 GitHub stars.
- Polis Protocol @yehudalevy-collab/polis-protocolLocal-first control plane for coding agents: routing, reservations, and guardrails.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.5 GitHub stars.
- md-feedback @yeominux/md-feedbackReview markdown plans before AI agents build. Annotate, gate-check, apply fixes, session handoffs.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.5 GitHub stars.
- Diffgate @srbsa/diffgateDeterministic guardrail your coding agent runs on itself: diff-scoped findings, zero LLM tokens.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.5 GitHub stars.
- Hashfile MCP @mrorigo/hashfileHash-anchored file editing with surgical precision over file editing and ironclad control of access4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.5 GitHub stars.
- ArchSteer @einvoice-dev1/archsteerArchitecture governance for AI agents: query the live model, get target patterns, check files.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.5 GitHub stars.
- Coding Standards @corbat-tech/coding-standardsAI coding standards that enforce production-grade code with DDD, SOLID, TDD guardrails.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.5 GitHub stars.
- Korg Ledger @new1direction/korg-ledgerVerifiable cognition: prove a session ledger is tamper-evident; audit & import any agent's logs.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.5 GitHub stars.
- nomArmy @rayson-tech/nomarmySandboxed coding workers, verified: tests rerun fresh and must fail with the change reverted.Not assessed yet.Runs on your machine, installed from npm.4 GitHub stars.
- AdversaryGate @sanflow10/adversary-gateFail-closed merge gate an agent calls before saying done: MERGE, BLOCK or INCONCLUSIVE.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.4 GitHub stars.
- Stuga @stuga/stugaAgent edits await review by default in a self-hosted workspace for documents and databases.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.4 GitHub stars.
- MemoryGuard @merakosiris/memoryguardLocal-first governed MCP memory backend for coding agents.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.4 GitHub stars.
- MetaGO Agent Harness @metago-ai/metagoDecision lock, anti-hallucination, meta-evolution and provenance for any MCP agent. 55 tools.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.4 GitHub stars.
- Aguara MCP @garagon/aguara-mcpSecurity scanner for AI agent skills and MCP servers4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.4 GitHub stars.
- Bulwark @bpolania/bulwarkAI agent governance: content scanning, audit logs, policy evaluation, session management.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.4 GitHub stars.
- Sabline @gowrishankar-infra/sablineRun code an AI wrote without handing it everything you can reach.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.4 GitHub stars.
- Tirith @eabz/tirithCoordination server for parallel coding agents: claims, contracts, notices, and memory notes.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from cargo.4 GitHub stars.
- Velaris @gowrishankar-infra/velarisWrite Velaris, audit what it can touch, and run it under an effect budget.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.4 GitHub stars.
