The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process infrastructure data from a live Infrahub instance, which serves as an external data source.
Ingestion points: Infrastructure node data is retrieved via the Infrahub MCP server using tools such as mcp__infrahub__get_nodes, mcp__infrahub__search_nodes, and mcp__infrahub__query_graphql, as described in SKILL.md and rules/mcp-tools.md.
Boundary markers: The instructions do not define specific boundary markers or "ignore" directives to prevent the agent from being influenced by instructions that might be embedded in the retrieved infrastructure data.
Capability inventory: The skill allows the use of the Bash and Read tools, and provides the capability to perform GraphQL mutations and propose changes to the live infrastructure state.
Sanitization: No explicit sanitization or validation of the JSON data returned from the MCP server is documented in the correlation and analysis rules.
[COMMAND_EXECUTION]: The skill provides several Python patterns for data correlation in rules/correlation-patterns.md. The agent is expected to use the allowed Bash tool to execute these logic patterns to process the data retrieved from Infrahub, creating a potential execution path for untrusted data.
[EXTERNAL_DOWNLOADS]: The skill references the official Infrahub documentation site (https://docs.infrahub.app) for further guidance on GraphQL and server interactions.

infrahub-analyzing-data