agent security agents

Tencent/AI-Infra-Guard

Agent

An automation procedure for deploying a Model Context Protocol (MCP) program from source code. MCP is a standard way for an AI client to connect to tools or data services.

Tencent/AI-Infra-Guard

Agent

A static security-audit agent for MCP projects. Static analysis examines source code without running it, focusing on vulnerabilities that could be reached through network inputs.

Tencent/AI-Infra-Guard

Agent

A security-review agent that checks vulnerability reports for real, reproducible threats and filters out false positives. A false positive is an alleged problem that is not actually exploitable or harmful in the stated environment.

Nasiko-Labs/nasiko

Agent

Agent "BUILD_WORKER" from Nasiko-Labs/nasiko, covering build worker, why a job queue?, database schema, job payload variants and lifecycle overview.

hashgraph-online/hol-guard

Agent

Review the evidence returned by the demo server.

pegasi-ai/reins

Agent

Cursor is an AI-powered code editor. ToolShield injects safety guidelines into Cursor's global user rules stored in its internal SQLite database.

pegasi-ai/reins

Agent

OpenClaw is an open-source AI coding agent. ToolShield injects safety guidelines into its workspace instruction file.

pegasi-ai/reins

Agent

OpenHands is an open-source AI-driven development platform. ToolShield injects safety guidelines as a microagent that is always loaded into the agent's system prompt.

Garudex-Labs/caracal

Agent Claude Code

Invoke ONLY when explicitly asked for an outside-in usability and adoption assessment of Caracal from a real external engineer's perspective. This agent evaluates: Console UX, SDK ergonomics, CLI experience, documentation clarity, deployment workflows, provider/gateway setup, policy model, app/resource modeling…

Garudex-Labs/caracal

Agent Claude Code

Invoke ONLY when explicitly asked for a production-readiness audit or hardening of the Caracal OSS platform: reliability, stability, recoverability, scalability, observability, performance, operational simplicity, deployment, upgrade safety, and maintainability. Finds production gaps, explains root causes, designs…

Garudex-Labs/caracal

Agent Claude Code

Use this agent when you need to prepare, validate, and publish a production release for a software project. This agent is designed for release engineering tasks including: preparing a new version release (e.g., 'Cut release v0.2.0 or v0.2.0-rc.1'), validating release readiness across all dimensions (code, packages…

Szesnasty/ai-protector

Agent

Folder: 01-agents-pre-tool-enforcement/ | Commit: c99c3be.

trustabl/trustabl

Agent Claude Code

Reads files and grep'd content, no writes.

trustabl/trustabl

Agent Claude Code

Email inbox search specialist, takes in appropriate context for the email and a goal of what to search for. Return the answer to the question.

trustabl/trustabl

Agent Claude Code

Deterministic-fixture subagent.

OpenLatch/saferskills

Agent

Pinned known-good config shapes the 8 writers target, with the official source + retrieval date. The MEDIUM/LOW (volatile) surfaces carry a live-verification gate: their checkbox stays ☐ until a maintainer installs the entry against a real running agent and confirms it loads, then ticks it here. A surface that cannot…

Diplomat-ai/diplomat-agent

Agent

Reviews Python code for unguarded AI agent tool calls — database writes, API calls, payments, emails, shell commands — that have no validation, rate limiting, or confirmation. Use this when reviewing agent code, after accepting Copilot suggestions, or before committing changes.

GeckoVision/gecko-surf

Agent Claude Code

defi-engineer.md, solana-architect.md, and solana-qa-engineer.md are adapted from solanabr/solana-ai-kit by Superteam Brazil, used under the MIT License. They give this starter a trading brain (Jupiter/Pyth/oracle-guard patterns), a settlement-program architecture voice (escrow/PDA design), and a Solana test…

GeckoVision/gecko-surf

Agent

API-provider onboarding specialist. Takes a provider's API (an OpenAPI URL, a docs page, or a pasted spec) and makes its FULL surface agent-ready — comprehend it with gecko into first-call-correct tools, emit the agent-native breadcrumbs (llms.txt / x-gecko / gecko.json), serve it over Streamable-HTTP MCP with a…

GeckoVision/gecko-surf

Agent

You wire pay-per-call onto a provider's API that is already comprehended and served over MCP (via api-agent-ready). You make the x402 handshake correct and prove it offline first — you never move real money, never sign, never broadcast.

adeelahmad/agentrc

Agent

Cross-sprint learning, curated at each retrospective. Entries are advisory and never relax an invariant. First agentic-agile session in this repo (Sprint 1); seeded from the 2026-07-02/03 docs+examples+spec audit that produced this work order.

tiagosilva07/zyrax-guard

Agent Claude Code

Use PROACTIVELY when starting a new feature, service, or MVP, or when a design decision needs to be made before code is written. Designs system architecture, data flow, API contracts, database schema, and a minimal-but-scalable implementation plan. Invoke for "design", "architecture", "how should we structure", "new…

tiagosilva07/zyrax-guard

Agent Claude Code

Use when something is broken, failing, flaky, or behaving unexpectedly, and the cause isn't obvious. Reproduces the issue, traces the real root cause, explains why it fails, surfaces hidden edge cases, and proposes the most robust fix. Invoke for "bug", "error", "crash", "failing test", "works locally but not in…

tiagosilva07/zyrax-guard

Agent Claude Code

Use to prepare an app for production deployment or improve its operational setup. Designs deployment architecture, CI/CD pipelines, containerization, monitoring and logging, and reliability/scaling. Invoke for "deploy", "CI/CD", "Docker", "Kubernetes", "pipeline", "monitoring", "logging", "infrastructure", "uptime"…