I'm seeking any feedback from the community on an open-source tool that I've been working on. It's essentially a tool that lets you create governance guardrails and insights for Claude Code.
* Full audit trail - every tool call logged with matched rule and reason
* Defense in depth - extendable policy layers: 9+ deterministic evaluators and an optional semantic layer (another LLM) as an evaluator; PII detection, config integrity checks, and injection detection
* Agent memory view - see what your agent sees
* Coverage and alignment - built-in compliance tags; add your own
I'd love any feedback on whether this can be useful for your workflow? What features do you find interesting? What's missing?
Hi HN,
I'm seeking any feedback from the community on an open-source tool that I've been working on. It's essentially a tool that lets you create governance guardrails and insights for Claude Code.
Here are some of the features:
* No data leaves your machine.
* Every policy and rule is controlled by you.
* Live dashboard - every decision logged locally
* Budget limits - usage patterns, cost caps, token and action limits
* Full audit trail - every tool call logged with matched rule and reason
* Defense in depth - extendable policy layers: 9+ deterministic evaluators and an optional semantic layer (another LLM) as an evaluator; PII detection, config integrity checks, and injection detection
* Agent memory view - see what your agent sees
* Coverage and alignment - built-in compliance tags; add your own
I'd love any feedback on whether this can be useful for your workflow? What features do you find interesting? What's missing?
Kind regards, Mo