daily-paper-generator
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted paper titles and abstracts from external sources, which could contain instructions intended to manipulate the agent's behavior.
- Ingestion points: Research paper abstracts and titles are retrieved from the arXiv and bioRxiv APIs via the
scripts/arxiv_search.pyscript. - Boundary markers: The skill's instructions in
SKILL.mddo not include explicit delimiters or instructions to treat the external content as non-executable data, which may allow the agent to follow instructions embedded in a paper's abstract. - Capability inventory: The agent has the capability to write the generated summaries and findings to the local file system in the
daily paper/directory. - Sanitization: No security-specific sanitization is applied to the fetched text before it is presented to the agent for processing and summarization.
Audit Metadata