spark-cli-knowledge-sharing
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill utilizes the
spark shareandspark share-taskcommands to transmit user-generated insights to an external repository. To mitigate security risks, the instructions include explicit rules forbidding the inclusion of source code, file contents, API keys, or credentials in shared data. - [COMMAND_EXECUTION]: The skill employs the
Bashtool to run a CLI binary namedspark. These commands are used for session management, knowledge retrieval, and feedback loops. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from an external knowledge base that may contain untrusted content.
- Ingestion points: Output from the
spark querycommand (SKILL.md). - Boundary markers: Absent; no delimiters or instructions to ignore embedded commands in the retrieved data are provided.
- Capability inventory: Access to the
Bashtool across multiple commands (SKILL.md). - Sanitization: Absent; the skill does not specify any validation or filtering of the content returned by the external service.
Audit Metadata