splunk-search
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation indicates the use of
npxto download and execute thesplunk-mcppackage from the NPM registry at runtime. The use of an unpinned and unverified package from a public registry introduces a dependency on external code that could be modified without user knowledge.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes and displays external data retrieved from Splunk logs, which serves as an injection surface for malicious instructions.\n - Ingestion points: Search results retrieved through the
search_oneshotandsearch_exporttools in SKILL.md.\n - Boundary markers: Absent. The skill does not define specific delimiters or instructions to prevent the agent from following commands embedded in log data.\n
- Capability inventory: The skill allows executing SPL queries and exporting data to the filesystem.\n
- Sanitization: While the documentation mentions sanitizing sensitive fields, it does not describe mechanisms to escape or filter instructions to prevent prompt injection.
Audit Metadata