ai-pdf-filler-cli

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the simplicity-cli command to perform PDF operations. It includes instructions for installing the CLI tool if it's missing using uv or pip.
  • [EXTERNAL_DOWNLOADS]: The skill encourages downloading and installing the ai-pdf-filler package from external repositories (PyPI/uv) to enable its functionality. This package is associated with the skill's author.
  • [CREDENTIALS_UNSAFE]: The skill instructs users to provide an API key as a command-line argument (--api-key "<api_key>"). While using placeholders, this pattern can lead to sensitive keys being stored in shell history or process listings. It also mentions a local configuration file ~/.config/simplicity-cli/config.json for storing the key.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. * Ingestion points: External data enters the agent context via the --form-url, --source-url, --context, and --instructions parameters in SKILL.md. * Boundary markers: There are no explicit boundary markers or instructions to the LLM to ignore embedded commands within the source documents or context strings. * Capability inventory: The skill has the capability to execute shell commands (simplicity-cli) and perform network operations (fetching remote PDFs). * Sanitization: There is no evidence of sanitization or validation of the content provided in the context or fetched from URLs before it is processed by the CLI tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 09:00 AM