The Agent Skills Directory

[SAFE]: No malicious code, prompt injections, or unauthorized network operations were identified. The skill follows best practices, providing clear documentation and unit tests for its processing logic.- [INDIRECT_PROMPT_INJECTION]: The skill represents an attack surface for indirect prompt injection because it ingests untrusted external content (claims and source passages). However, the capability to exploit this is negligible as the script only generates Markdown text.
Ingestion points: Factual assertions and source passages are loaded from claim.json and sources.json in scripts/verify.py.
Boundary markers: Claim text and extracted passages are enclosed in double quotes within the resulting Markdown report.
Capability inventory: The script can write the generated report to a local file path using the --output argument.
Sanitization: The script performs basic sanitization by replacing newline characters with spaces in source passages.

fact-checker