The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting data from external sources and using it to guide remediation actions.\n
Ingestion points: Workflow 1 in SKILL.md retrieves alert details, incident assessments, table schemas, and pipeline query logs from the Monte Carlo platform.\n
Boundary markers: The instructions do not specify the use of delimiters or boundary markers when processing these external inputs to separate data from instructions.\n
Capability inventory: The agent is granted shell access, the ability to execute Monte Carlo MCP tools, and authorization to trigger external remediation tools such as CLI utilities or REST APIs.\n
Sanitization: No explicit sanitization or validation logic is defined for the external data ingested during the investigation phase.\n- [COMMAND_EXECUTION]: The skill facilitates the discovery and execution of various CLI tools and shell commands to perform remediation tasks.\n
Evidence: Workflow 2 instructs the agent to use the which command to check for the availability of tools like gh, dbt, and airflow in the shell environment.\n
Evidence: Workflow 3 describes executing remediation actions, including pipeline restarts and database job reruns, via these discovered tools.\n- [EXTERNAL_DOWNLOADS]: The skill utilizes network-capable tools to interact with external services during the remediation process.\n
Evidence: Workflow 2 mentions that REST APIs can be reached via curl to perform remediation actions when MCP tools or CLI utilities are unavailable.

monte-carlo-remediation