The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to download and run the splunk-mcp package from the public npm registry at runtime. This introduces a dependency on code from an unverified external source.
[REMOTE_CODE_EXECUTION]: Executing npx splunk-mcp allows the code from the downloaded npm package to run on the local host system.
[COMMAND_EXECUTION]: The skill facilitates the execution of SPL queries on a Splunk server. If user input is not properly sanitized before being included in these queries, it could lead to SPL injection.
[PROMPT_INJECTION]: The skill ingests untrusted log data from Splunk, which creates a surface for Indirect Prompt Injection if the data contains malicious instructions for the agent.
Ingestion points: Search results retrieved from the search_oneshot and search_export tools in SKILL.md.
Boundary markers: No delimiters or markers are defined to isolate external Splunk data from the rest of the agent's prompt context.
Capability inventory: The skill has the ability to execute searches and export results to the filesystem.
Sanitization: The skill documentation mentions automatic sanitization of sensitive fields, but no implementation details are provided to verify the effectiveness against prompt injection attacks.

splunk-search