algo-seo-tfidf

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions do not contain any attempts to override agent behavior, bypass safety filters, or extract system prompts. The language is strictly instructional and related to the SEO domain.- [DATA_EXFILTRATION]: No hardcoded credentials, sensitive file paths, or network operations were found. The Python script operates locally on data provided via a JSON input file.- [REMOTE_CODE_EXECUTION]: No external package installations or remote script executions (e.g., curl|bash) are present. The script uses only the Python standard library.- [COMMAND_EXECUTION]: The script scripts/tfidf.py implements mathematical logic for scoring and does not use dangerous functions like eval(), exec(), or subprocess to execute arbitrary commands.- [OBFUSCATION]: No hidden content, encoded strings, or homoglyph substitutions were detected in the instructions or the source code.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process document corpora which could contain untrusted data.
  • Ingestion points: The skill reads document text from JSON files in the scripts/tfidf.py script.
  • Boundary markers: None explicitly defined in the prompts, though data is expected in a structured JSON format.
  • Capability inventory: The skill's capabilities are restricted to text tokenization and mathematical scoring; it lacks file-writing, network access, or system command capabilities.
  • Sanitization: The script tokenizes input using alphanumeric regex and lowercasing, which effectively treats the data as passive text strings rather than instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 07:10 AM