The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes data from external sources including Monte Carlo alert contexts, troubleshooting agent logs, and data warehouse metadata. If these sources contain malicious strings, they could influence the agent's actions.
Ingestion points: Data is ingested via get_alerts, run_troubleshooting_agent, get_table, and get_queries_for_table as described in SKILL.md and references/patterns.md.
Boundary markers: The instructions lack explicit requirements for using delimiters or boundary markers to isolate retrieved data from system instructions.
Capability inventory: The skill utilizes shell access for gh, dbt, airflow, and curl, along with MCP tools that can trigger pipeline runs and modify alert statuses.
Sanitization: There are no explicit requirements in the skill files to sanitize or validate the content of retrieved alerts or metadata before processing.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to discover and execute various CLI tools such as gh, dbt, airflow, and curl through shell access. While references/safety.md includes safety rails requiring user confirmation for destructive actions, the core functionality relies on executing powerful commands in the environment.

monte-carlo-remediation