aicordon (Picket) is a lightweight, local security tool that detects indirect prompt injection attempts using fast rule-based signature matching. It requires no machine learning models or network connectivity, making it suitable for integration into LLM applications, agents, and guardrail pipelines where low latency and offline operation are important.
In the Developer Tools space, aicordon takes a focused approach. It focuses on detecting indirect prompt injection attacks in LLM inputs without relying on external models or network calls. It is built as an open-source project for developers. The project is open source (Apache-2.0). aicordon is available on the web and the command line.
Behind aicordon is AICordon, and it first shipped in 2026. The project is developed in the open on GitHub with 34 commits in the last 90 days. Among its 3 catalogued features are Prompt Injection Detection, Local Signature Matching, and Rule-based Analysis.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match