pp-openrouter-image

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs replaying playbook steps with Playbook.slots_resolved substitutions (which include a "token" field) so an agent may be required to insert live token/API-key values verbatim into generated commands, creating an exfiltration risk despite recommending env-var auth elsewhere.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes an “automatic learning/recall” loop where the CLI reads the user’s question string via openrouter-image-pp-cli recall "<user's question>" --agent, and then reads/acts on locally stored learning content that can be influenced through that user input.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill requires installing and running a remote CLI fetched at runtime (e.g., "github.com/mvanhorn/printing-press-library/library/ai/openrouter-image/cmd/openrouter-image-pp-cli" via go install/npx), which fetches and executes remote code and is a required dependency for the skill.