very-long-text-summarization
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and process large volumes of untrusted data from external sources such as documents and URLs. It lacks explicit boundary markers or instructions to distinguish data from instructions during the extraction and synthesis phases.\n
- Ingestion points: External text documents provided via the
[file-path-or-url]argument inSKILL.md.\n - Boundary markers: Absent. The extraction and synthesis templates provided in
SKILL.mddo not include instructions for the agent to ignore or delimit potentially malicious instructions embedded within the source text.\n - Capability inventory: The skill allows use of
Bash,Write,Edit, andReadtools, which provides a significant capability surface if the agent follows instructions found within a document.\n - Sanitization: Absent. No content filtering, escaping, or validation of the input text is described in the skill logic.
Audit Metadata