dependency-analysis

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.70). The prompt includes an out-of-scope "SELF-EVOLUTION TELEMETRY" section that instructs the agent to search the user's home for ~/.claude/projects and append telemetry to skill-telemetry.md — an explicit side-effect/exfiltration-style action unrelated to pure dependency analysis and therefore a hidden/deceptive instruction outside the skill's stated purpose.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The skill's workflow explicitly requires cross-referencing dependencies against public package registries and vulnerability/advisory databases (e.g., npm/PyPI, CVE database, Snyk DB, GitHub advisories) — open third-party sources whose package metadata/vulnerability reports the agent must read and which can materially influence update and remediation actions.