well-log-evaluation
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data files (LAS and DLIS formats) which are common in the energy industry. While no malicious activity is present in the skill itself, processing untrusted data files creates an attack surface where embedded adversarial content could theoretically influence the agent's behavior during analysis.
- Ingestion points: External files are loaded using
lasio.read()anddlisio.dlis.load()inSKILL.md(Stage 1). - Boundary markers: None identified in the provided code snippets or instructions.
- Capability inventory: The skill utilizes file system read access for data loading and standard data processing libraries (NumPy, Pandas, Welley, Petropy) for formation evaluation and visualization.
- Sanitization: No specific sanitization or validation logic for the content of the log files is described beyond standard parsing provided by the libraries.
Audit Metadata