verify-artifact
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to ingest and analyze external artifacts (e.g., documents, code, data files) to verify their integrity. These files constitute untrusted input that could contain adversarial instructions. \n
- Ingestion points: Derived artifacts (PDF, JSON, CSV, code) described in the verification workflows. \n
- Boundary markers: The skill does not provide specific instructions or delimiters for the agent to isolate the artifact content from its internal execution logic. \n
- Capability inventory: The skill has access to
Read,Bash,Grep,Glob, andWritetools. \n - Sanitization: No explicit sanitization or filtering of the artifact data is performed before parsing or verification. \n- [COMMAND_EXECUTION]: The skill instructions advise using the
Bashtool to run system utilities for file validation (e.g.,pdfinfo,pdftotext,unzip -t) and to execute build scripts. This usage is consistent with the skill's primary objective of artifact verification and does not involve the execution of suspicious or hidden commands.
Audit Metadata