pp-juneoven

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill tells agents to replay playbook steps by substituting Playbook.slots_resolved entries (which can contain "token"/live secret values) directly into commands and allows inline playbook JSON, forcing the LLM to emit those slot values verbatim when constructing commands — an exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). Source: June Oven Printing Press CLI skill, runtime path “Agent Mode” where the agent executes juneoven-pp-cli recall "<user's question>" --agent (and also teach --query "<user's question>" ...), meaning outsider-authored free text from the user is directly ingested by the CLI’s recall/decision logic at runtime.

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 remote packages that fetch and execute code at runtime (e.g., go install github.com/mvanhorn/printing-press-library/library/devices/juneoven/cmd/juneoven-pp-cli@latest), so the agent can pull and run unverified external code.