jev-capability-advisor
Warn
Audited by Snyk on Nov 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required workflow ingests skill and MCP tool descriptions from the host's catalog, and enrichment can read frontmatter of already-installed third-party skills, but an outsider cannot post into this stream and the agent does not select the items. Exposure therefore requires a prior install or publish step, which is the low "publish poison, cannot force delivery" case.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (low risk: 0.30). The skill has the agent run an
installthat persistently registers a hook in user or project Codex/Claude Code configuration, which changes persistent agent-host state, though it is consent-gated, needs no sudo, and does not touch system files, accounts, services, or SSH config.
Issues (2)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata