illo
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads character packs and visual assets from the author's public GitHub repositories (
tmchow/illo-charactersandtmchow/agent-skills). These operations are essential to the skill's functionality and include integrity verification via SHA256 checksums for asset repair. - [PROMPT_INJECTION]: While the skill processes untrusted input from external URLs and article content (indirect prompt injection surface), it includes specific architectural guardrails to mitigate this. It instructs the agent to ignore any commands found within data files and enforces a "thesis locking" step to ensure the output is derived from the agent's analysis rather than embedded instructions.
- [COMMAND_EXECUTION]: The skill executes a bundled Python engine (
illo.py) and an asset-repair bash script. These executions are scoped to the skill's internal logic, primarily handling API communication and filesystem maintenance within the skill's own directory and the user's local config folder.
Audit Metadata