octocode-research

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the agent’s LLM context via tool responses returned by the server—e.g., ghGetFileContent/ghSearchCode/npmSearch fetch GitHub/npm content at runtime, and the server parses and returns that text (including content[0].text) as data/hints in the /tools/call/:toolName response, which the agent then reads into its context.