internal-comms
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) as its primary purpose is to aggregate and process data from various communication channels.\n
- Ingestion points: The agent is instructed to gather content from Slack messages, Google Drive documents, emails, and calendar events as specified in the examples directory.\n
- Boundary markers: The instructions lack delimiters or explicit warnings to the agent to disregard instructions found within the retrieved data.\n
- Capability inventory: The agent reads sensitive internal communications to generate summaries; if the source data contains malicious prompts, they could alter the agent's output or actions.\n
- Sanitization: No sanitization or validation logic is provided to filter the content before it is used for drafting communications.\n- [NO_CODE]: The skill package contains only markdown instruction files and does not include any executable scripts or binary files.
Audit Metadata