octocode

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

• Third-party content exposure detected (high risk: 0.75). The required workflow runs octocode commands that can fetch and read outsider-authored public web/GitHub content at runtime (e.g., octocode cat/grep/ls/find/pr/repo on owner/repo or URLs), which the agent then ingests as readable text into the LLM context.