Human-in-the-loop execution for LLM agents
python linux cli security devops automation mcp sandbox sysadmin python3 developer-tools human-in-the-loop llm llm-agents agent-safety supervised-execution
-
Updated
Jan 11, 2026 - Python
#
agent-safety
Here are 23 public repositories matching this topic...
Language: All
Filter by language
Sort: Most stars
Sort options
Guardrails for LLMs: detect and block hallucinated tool calls to improve safety and reliability.
middleware machine-learning ai language-models ai-safety prompt-engineering llms toolformer hallucination-detection tool-calling agent-safety
-
Updated
Jul 18, 2025 - Go
π‘οΈ Safe AI Agents through Action Classifier
python agent ai hackathon mcp hackathons agents ai-safety ai-agents tool-use ai-agent ai-evaluation llms generative-ai gen-ai llm-benchmarking agent-tools agentic-ai agent-safety tool-orchestration
-
Updated
Feb 17, 2026 - Python
π‘οΈ Open-source safety guardrail for AI agent tool calls. <2ms, zero dependencies.
-
Updated
Feb 15, 2026 - TypeScript
The missing safety layer for AI Agents. Adaptive High-Friction Guardrails (Time-locks, Biometrics) for critical operations to prevent catastrophic errors.
-
Updated
Jan 28, 2026 - TypeScript
A runtime authorization layer for LLM tool calls policy, approval, audit logs.
python agent security authorization developer-tools human-in-the-loop policy-engine jsonl runtime-security audit-logging guardrails llm agent-safety
-
Updated
Feb 6, 2026 - Python
Runtime detector for reward hacking and misalignment in LLM agents (89.7% F1 on 5,391 trajectories).
nlp machine-learning monitoring deep-learning transformers pytorch alignment ai-safety fastapi huggingface streamlit distilbert llm rlhf llm-agents agent-safety reward-hacking misalignment
-
Updated
Dec 11, 2025 - Python
ETHICS.md β A statement of ethical principles for AI agents. Drop it in your repo root.
readme developer-tools ai-safety ethics ai-agents claude ai-ethics ai-alignment responsible-ai llm prompt-injection agent-safety ethics-md
-
Updated
Feb 19, 2026
Safety-first agentic toolkit: 10 packages for collapse detection, governance, and reproducible runs.
benchmark time-series simulation reliability observability governance ethics anomaly-detection mlops agent-safety
-
Updated
Dec 9, 2025 - Python
A2A version of Agent Action Guard: Safe AI Agents through Action Classifier
python agent ai hackathon hackathons agents ai-safety ai-agents hackathon-project ai-research ai-ethics ai-agent ai-monitoring ai-evaluation llms generative-ai gen-ai llms-benchmarking agentic-ai agent-safety
-
Updated
Dec 14, 2025 - Python
An open-source engineering blueprint for defining and designing the core capabilities, boundaries, and ethics of any AI agent.
protocol specification standard autonomous-agents dev-tools agp ai-ethics agent-framework ai-agent agent-design llm llm-agents agent-architecture agent-safety
-
Updated
Sep 6, 2025
A hierarchical AI safety architecture with asymmetric supervisory control.
architecture alignment governance ai-safety interpretability oversight robust-ai hierarchical-control ai-governance auditability agent-safety
-
Updated
Jan 4, 2026
Energy based legality gating SDK for AI reasoning. Predicts, repairs, and audits collapse before it happens; reduces hallucinations and provides numeric audit logs.
middleware reliability audit compliance observability control-theory ai-safety llm reasoning-language-models agent-safety
-
Updated
Oct 25, 2025
Canonical texts and implementation primitives for the Safe Superintelligence Framework (v1.2.1): Constitution, Minimum Rescue Protocol, system prompt, decision matrix.
ai-safety risk-management ai-alignment responsible-ai ai-governance system-prompt auditability agent-safety minimum-rescue
-
Updated
Jan 3, 2026
A security-first control plane for autonomous AI code agents: sandboxed execution, hash grounding, diff validation, verification, and full auditability.
python mcp devtools developer-tools observability autonomous-agents ai-agents fastapi secure-ai llm ai-infrastructure agent-safety
-
Updated
Jan 22, 2026 - Python
A protocol engine for governing AI agent workflows through gated checkpoints and immutable audit trails.
-
Updated
Feb 6, 2026 - Python
Semantic differential protection layer for AI agents. The semantic analogue of differential protection (RCD) in electrical systems.
security monitoring semantics trajectory anomaly-detection ai-agents runtime-monitoring guardrails llm agent-safety behavior-monitoring
-
Updated
Feb 19, 2026 - Python
Prompt Injection Firewall for AI agents. 113 detection patterns, 14 threat categories, zero dependencies. Protects against fake authority, command injection, memory poisoning, skill malware, and more.
python cli security pattern-matching firewall zero-dependency ai-safety ai-agent prompt-injection llm-security claude-code agent-safety
-
Updated
Feb 19, 2026 - Python
Production-ready safety framework preventing identity fusion, synthetic intimacy, and unbounded behavior in AI agent systems. Machine-readable contracts and verse-lang primitives for immediate deployment.
safeguarding autonomous-agents ai-safety ai-agents ai-ethics policy-as-code machine-readable prompt-injection llm-safety agent-safety agent-governance verse-ality relational-safety synthetic-intimacy boundary-engine identity-fusion safety-framework moltbook safety-contracts containment-protocols
-
Updated
Feb 2, 2026 - Python
π‘οΈ Safeguard AI agents from harmful actions with A2A-Agent-Action-Guard, ensuring safe tool usage through effective action classification.
python agent ai hackathon agents ai-safety ai-agents hackathon-project ai-research ai-ethics ai-agent ai-monitoring ai-evaluation generative-ai gen-ai llms-benchmarking agentic-ai agent-safety
-
Updated
Feb 20, 2026 - Python
Improve this page
Add a description, image, and links to the agent-safety topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the agent-safety topic, visit your repo's landing page and select "manage topics."