mthamil107/prompt-shield

enhancement: bring Anthropic wrapper to parity with OpenAI wrapper's tool_result_mode (post-#34) 1 day ago

enhancement help wanted

15

Prompt-injection firewall for LLM applications — 33 input detectors, 9 output scanners, federated ed25519-signed threat-intel feed. Apache 2.0, 1040 tests, F1 96.0% with 0% false positives. Docker, GitHub Action, LangChain/LlamaIndex/CrewAI integrations.

Python

#agentic-ai#ai-safety#benchmarks#crewai#fastapi#guardrails#langchain#llamaindex#llm-firewall#llm-security#mcp#owasp#pii-detection#prompt-injection#prompt-shield#python#red-teaming#self-learning#threat-intelligence#vector-similarity

bug: ToolResultGuard.ascan cache mutation race under concurrent callers 5 days ago

bug help wanted

mthamil107/prompt-shield

15

Prompt-injection firewall for LLM applications — 33 input detectors, 9 output scanners, federated ed25519-signed threat-intel feed. Apache 2.0, 1040 tests, F1 96.0% with 0% false positives. Docker, GitHub Action, LangChain/LlamaIndex/CrewAI integrations.

Python

#agentic-ai#ai-safety#benchmarks#crewai#fastapi#guardrails#langchain#llamaindex#llm-firewall#llm-security#mcp#owasp#pii-detection#prompt-injection#prompt-shield#python#red-teaming#self-learning#threat-intelligence#vector-similarity

v0.8.0: pydantic-ai — add scan_tool_result primitives 5 days ago

enhancement help wanted

mthamil107/prompt-shield

15

Prompt-injection firewall for LLM applications — 33 input detectors, 9 output scanners, federated ed25519-signed threat-intel feed. Apache 2.0, 1040 tests, F1 96.0% with 0% false positives. Docker, GitHub Action, LangChain/LlamaIndex/CrewAI integrations.

Python

#agentic-ai#ai-safety#benchmarks#crewai#fastapi#guardrails#langchain#llamaindex#llm-firewall#llm-security#mcp#owasp#pii-detection#prompt-injection#prompt-shield#python#red-teaming#self-learning#threat-intelligence#vector-similarity

v0.8.0: CrewAI — add scan_tool_result method 6 days ago

enhancement help wanted

mthamil107/prompt-shield

15

Prompt-injection firewall for LLM applications — 33 input detectors, 9 output scanners, federated ed25519-signed threat-intel feed. Apache 2.0, 1040 tests, F1 96.0% with 0% false positives. Docker, GitHub Action, LangChain/LlamaIndex/CrewAI integrations.

Python

#agentic-ai#ai-safety#benchmarks#crewai#fastapi#guardrails#langchain#llamaindex#llm-firewall#llm-security#mcp#owasp#pii-detection#prompt-injection#prompt-shield#python#red-teaming#self-learning#threat-intelligence#vector-similarity

v0.8.0: OpenAI wrapper — route role="tool" messages through ToolResultGuard 6 days ago

enhancement help wanted

mthamil107/prompt-shield

15

Prompt-injection firewall for LLM applications — 33 input detectors, 9 output scanners, federated ed25519-signed threat-intel feed. Apache 2.0, 1040 tests, F1 96.0% with 0% false positives. Docker, GitHub Action, LangChain/LlamaIndex/CrewAI integrations.

Python

#agentic-ai#ai-safety#benchmarks#crewai#fastapi#guardrails#langchain#llamaindex#llm-firewall#llm-security#mcp#owasp#pii-detection#prompt-injection#prompt-shield#python#red-teaming#self-learning#threat-intelligence#vector-similarity

Open Issues Need Help