momentum-trend
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from CSV files via the
momentum_trend_diagnostics.pyscript, creating a potential surface for indirect prompt injection. - Ingestion points: The script
scripts/momentum_trend_diagnostics.pyreads data from an input CSV file provided as a command-line argument. - Boundary markers: The skill does not provide specific instructions or delimiters for the agent to safely isolate or ignore potentially malicious instructions embedded within the CSV data.
- Capability inventory: The diagnostic script includes capabilities for reading local files and writing JSON artifacts to the file system.
- Sanitization: While the script validates the presence of required column headers, it does not sanitize the actual content of the data rows for embedded prompt instructions or malicious payloads.
Audit Metadata