research-idea-discovery
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill's workflow involves ingesting and processing untrusted external data, which creates a surface for indirect prompt injection attacks.
- Ingestion points: The skill is instructed to read from
papers/,literature/,zotero/,literature-review/, and user notes as defined inSKILL.mdandcatalog.json. - Boundary markers: The instructions lack explicit delimiters or specific directives for the agent to ignore or isolate instructions that may be embedded within the research data it processes.
- Capability inventory: The skill utilizes Python scripts (
scripts/init_ideation_pack.pyandscripts/validate_ideation_pack.py) to perform file system read and write operations on the local workspace. - Sanitization: No sanitization or data validation measures are specified to handle potentially malicious content in the external research inputs.
Audit Metadata