pdf-statement-parser

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill handles sensitive financial data and personally identifiable information (PII). It reduces the risk of data exposure by mandating the use of account masks (last 4 digits) and strictly forbidding the extraction or storage of full account numbers.
  • [COMMAND_EXECUTION]: The workflow involves using the platform's 'Read' tool to access local PDF documents at user-provided paths for data extraction. This filesystem access is limited to the skill's intended document-parsing function.
  • [PROMPT_INJECTION]: The skill processes untrusted content from external PDF files, which represents an indirect prompt injection surface. Ingestion points: PDF documents read via the 'Read' tool. Boundary markers: Absent for ingested content. Capability inventory: Filesystem read access via the 'Read' tool. Sanitization: Absent. The risk is mitigated by a strict JSON output contract and 'Guardrails' that prevent the agent from inventing values or following embedded instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 02:35 AM