The Agent Skills Directory

[METADATA_POISONING]: There is a discrepancy between the author identified in the file metadata ('netclaw') and the expected author context ('automateyournetwork').
[EXTERNAL_DOWNLOADS]: The skill uses npx to run the splunk-mcp package, which requires downloading the package from the npm registry.
[INDIRECT_PROMPT_INJECTION]: The skill processes results from Splunk searches, which could contain untrusted data capable of influencing the agent's behavior if malicious instructions are embedded in log data.
Ingestion points: Data returned by the run_saved_search and get_saved_searches tools in SKILL.md.
Boundary markers: No delimiters or instructions to ignore embedded content are defined.
Capability inventory: Provides tools to list and execute search commands within the Splunk environment.
Sanitization: No specific sanitization or filtering of search result content is specified.

splunk-saved