pdf-reader

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and process untrusted external data (PDF files).
  • Ingestion points: Processes user-provided PDF documents for extraction and analysis as described in SKILL.md.
  • Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from obeying instructions embedded within the PDF content itself.
  • Capability inventory: The agent is instructed to output extracted data into structured formats like JSON and CSV, which could be leveraged if an attacker embeds instructions in a PDF to exfiltrate or manipulate data.
  • Sanitization: No content sanitization or validation logic is specified to mitigate the risk of instructions hidden in document text, metadata, or form fields.
  • [NO_CODE]: The skill contains only natural language instructions and metadata. No executable scripts, shell commands, or binary files are included in the skill package.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 08:44 AM