teach
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from user-specified files.
- Ingestion points: The skill reads content from arbitrary files in the codebase as specified by the user (e.g.,
@models/staging/stg_orders.sql) in theWorkflowsection ofSKILL.md. - Boundary markers: There are no explicit instructions or delimiters used to separate the file content from the agent's instructions, nor are there warnings to ignore executable instructions within the target files.
- Capability inventory: The skill utilizes the
training_savetool to persist data to the project's configuration/training store, as defined in step 5 of theWorkflow. - Sanitization: There is no mention of sanitization, validation, or escaping of the file content before it is analyzed and passed to the
training_savetool.
Audit Metadata