tao-generate-od-defects
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script "scripts/generate_od_defects.py" uses "subprocess.run" to execute internal processing tools such as "torchrun". This is the primary function of the skill and is performed within a container environment. The commands are constructed using lists rather than strings, which mitigates shell injection risks, and input variables are validated for existence on the file system before being used as arguments.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from a generation plan file which could be influenced by untrusted inputs in a workflow chain.
- Ingestion points: "scripts/generate_od_defects.py" reads and parses generation data from "anomalygen_next_generation_plan.json".
- Boundary markers: No explicit delimiters or safety warnings are used to isolate instructions from data within the JSON fields.
- Capability inventory: The skill possesses the capability to execute shell commands via "subprocess.run" and perform file system writes.
- Sanitization: The "_normalize" function performs input validation by checking "dataset_id" against a strict alphanumeric regex and verifying the existence of all referenced file paths on the local file system.
Audit Metadata