omh-paper-learning
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external content from research papers and PDFs. It mitigates the risk of embedded instructions by enforcing strict boundary markers, a coverage ledger, and source_state tracking to ensure the agent distinguishes between original text and its own analysis.- [COMMAND_EXECUTION]: Uses framework-specific CLI tools (e.g.,
omh paper plan,omh paper progress) to manage the state of the paper learning session. These tools are used for logging progress and validating coverage within the expected operational environment.- [DATA_EXPOSURE]: Writes and reads session-specific data (e.g.,card.json,ledger.jsonl) within the$OMH_HOMEdirectory. This behavior is limited to the skill's functional requirement of persistence across sessions and does not involve accessing sensitive system or user credentials.
Audit Metadata