Trust nothing. Ship safely. — Skeptical-reading and prompt-injection defense skill for AI agents. Provenance tagging, red-flag patterns, refusal templates, and a read-only injection auditor. MIT.
GitHub repository with 30 stars and 5 forks.
Language: Shell
Topics: agent-safety, ai-agents, ai-security, claude-code, claude-code-skill, llm-security, mcp, mcp-security, plugin, prompt-injection
Trending score 0.45, activity score 0.00, stars gained +2, forks gained +0.
2026-06-05: 30 stars and 5 forks.
Trust nothing. Ship safely. — Skeptical-reading and prompt-injection defense skill for AI agents. Provenance tagging, red-flag patterns, refusal templates, and a read-only injection auditor. MIT.
GitHub repository with 30 stars and 5 forks.
Trending score: 0.45; stars gained: +2; forks gained: +0.
Language: Shell
Topics: agent-safety, ai-agents, ai-security, claude-code, claude-code-skill, llm-security
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
GitHub repository with 107,462 stars and 17,720 forks.
Trending score: 4.22; stars gained: +233; forks gained: +38.
Language: Shell
Claude Code Dedicated Development Harness - Achieving High-Quality Development Through an Autonomous Plan→Work→Review Cycle
GitHub repository with 2,617 stars and 252 forks.
Trending score: 3.94; stars gained: +147; forks gained: +10.
Language: Shell
Measuring frontier coding agents on original, long-horizon engineering tasks
GitHub repository with 617 stars and 33 forks.
Trending score: 2.87; stars gained: +34; forks gained: +2.
Language: Shell
Mirror official Codex desktop app installers into GitHub Releases.
GitHub repository with 378 stars and 20 forks.
Trending score: 2.70; stars gained: +33; forks gained: +3.
Language: Shell
Topics: codex-app, github-actions, installer-mirror, macos, microsoft-store, msix
Agent2Agent (A2A) is an open protocol enabling communication and interoperability between opaque agentic applications.
GitHub repository with 24,132 stars and 2,448 forks.
Trending score: 2.64; stars gained: +28; forks gained: +4.
Language: Shell
Topics: a2a, a2a-mcp, a2a-protocol, a2a-server, agents, generative-ai
GitHub repository with 2,657 stars and 373 forks.
Trending score: 2.39; stars gained: +15; forks gained: +2.
Language: Shell
Deterministic execution authorization for AI agents
GitHub repository with 12 stars and 3 forks.
Trending score: 0.50; stars gained: +1; forks gained: +1.
Language: TypeScript
Topics: agent-runtime, agent-safety, deterministic-systems, policy-engine, runtime-guardrails, execution-authorization
Trust nothing. Ship safely. — Skeptical-reading and prompt-injection defense skill for AI agents. Provenance tagging, red-flag patterns, refusal templates, and a read-only injection auditor. MIT.
GitHub repository with 30 stars and 5 forks.
Trending score: 0.45; stars gained: +2; forks gained: +0.
Language: Shell
Topics: agent-safety, ai-agents, ai-security, claude-code, claude-code-skill, llm-security
The open-source safety layer for AI agents — block unsafe tool calls, require approval, enforce budgets, audit, replay.
GitHub repository with 13 stars and 5 forks.
Trending score: 0.11; stars gained: +0; forks gained: +0.
Language: Rust
Topics: ai-agents, llm, mcp, python, rust, workflow
GitHub repository with 110 stars and 8 forks.
Trending score: 0.06; stars gained: +0; forks gained: +0.
Language: Python
Topics: agent-safety, ai-agents, audit, code-review, codex, developer-tools
Runtime detector for reward hacking and misalignment in LLM agents (89.7% F1 on 5,391 trajectories).
GitHub repository with 12 stars and 0 forks.
Trending score: 0.04; stars gained: +0; forks gained: +0.
Language: Python
Topics: agent-safety, ai-safety, distilbert, fastapi, huggingface, llm-agents
Execution control layer for AI agents — prevents duplicate or incorrect real-world actions under retries, uncertainty, and stale context.
GitHub repository with 6 stars and 2 forks.
Trending score: 0.04; stars gained: +0; forks gained: +0.
Language: Python
Topics: ai-agents, idempotency, reliability, agent, agent-infrastructure, agent-safety