learn
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user-editable log files (
out/learn/<slug>.md) which are used to resume learning sessions. - Ingestion points: The agent reads
out/learn/<slug>.mdat the start of every run to retrieve the syllabus, review queue, and verified sources. - Boundary markers: Absent. The instructions do not define delimiters or provide specific warnings to the model to ignore instructions found within the log file.
- Capability inventory: The skill has access to
WebSearchandWebFetchtools, as well as file system read/write access to theout/directory. It also has the ability to spin up sub-agents for grading. - Sanitization: Absent. There is no sanitization or validation of the log content before it is used to influence the agent's behavior, drive network searches, or determine the next teaching unit.
Audit Metadata