Skip to content

Add skillvet (pre-install static trust gate for agent skills)#27

Open
cognis-digital wants to merge 1 commit into
LLMSecurity:mainfrom
cognis-digital:add-skillvet
Open

Add skillvet (pre-install static trust gate for agent skills)#27
cognis-digital wants to merge 1 commit into
LLMSecurity:mainfrom
cognis-digital:add-skillvet

Conversation

@cognis-digital

Copy link
Copy Markdown

Adding skillvet to Tools & Frameworks.

skillvet is a static "trust gate" that vets a Claude Skill / MCP server / plugin before you install it — it reads the package (never executes it) and flags process execution, network egress, credential/filesystem access, install hooks, and obfuscated/dynamic-fetch code, each mapped to a MITRE ATLAS technique, then returns a TRUST / REVIEW / BLOCK verdict with a 0–100 trust score. Exit codes gate CI/pre-install.

It fills the pre-deployment static vetting niche (complementary to runtime/eval tools already listed). Open-source (COCL), stdlib-only, actively developed with a test suite. One tool per PR per CONTRIBUTING.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant