autocontext
Warn
Audited by Socket on Apr 17, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The stated purpose is coherent for a context-management skill, but the actual trust boundary is unacceptable: it requires an unverifiable external `autocontext` CLI, forwards Pinecone credentials to it, and sends data to an unspecified API instance. With missing provenance for the binary and undocumented network endpoints, the skill's footprint is not sufficiently trustworthy for its claimed publisher.
Confidence: 89%Severity: 86%
Audit Metadata