local-paper-index
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local PDF and markdown files, which could contain malicious instructions designed to influence the agent's behavior during indexing or subsequent search steps.
- Ingestion points: Reads PDF files from a user-specified directory and markdown files within the
assets/chunk-and-index.pyscript. - Boundary markers: The skill does not implement delimiters or explicit "ignore embedded instructions" warnings when processing the extracted text.
- Capability inventory: The skill has access to shell execution, local script execution, and file writing capabilities.
- Sanitization: Extracted text is chunked and written directly into the index YAML file without filtering or sanitizing the content for prompt injection patterns.
- [COMMAND_EXECUTION]: The skill executes multiple local scripts and system utilities to perform its tasks.
- Evidence: Calls
assets/extract-pdfs.sh,assets/chunk-and-index.py, andassets/warm-cache.shvia shell commands. It also uses common utilities likefind,wc,du, andmkdirto manage the file processing pipeline.
Audit Metadata