The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute gcx CLI commands. This is the primary mechanism for interacting with the Grafana Agent Observability stack and is appropriate for the skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted live traffic data from conversations to recommend security policies. While this presents an attack surface where malicious user input could influence the agent's recommendations, the risk is effectively mitigated by mandatory manual confirmation and a local drafting process.
Ingestion points: Live conversation data retrieved via gcx agento11y conversations search in Step 1.
Boundary markers: None explicitly defined for the processing of conversation text.
Capability inventory: Bash (CLI execution), Read (file system access), Write (drafting configurations), and Edit tools.
Sanitization: The skill does not explicitly sanitize the raw conversation strings during analysis, but requires an explicit "yes" from the developer before any resulting configuration is applied.

agento11y-prod-setup