review-skill-sflabs

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill instructs the agent to read all files under the skill directory into context and to include "evidence", diffs, and substituted findings in the rendered report and proposed changes, which can require reproducing secret values verbatim if they exist in those files.

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

• Third-party content exposure detected (medium risk: 0.30). In Phase 1/2 the runtime reads the candidate skill’s on-disk SKILL.md and sibling files under references/, workflows/, and scripts/ from the user-provided <path>, so outsider-authored free text can be ingested without selecting a specific item first.