The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill uses the !command`` syntax in SKILL.md to automatically provide a project file overview and display the .infrahub.yml configuration when the skill is loaded. These commands (find and cat) are used for legitimate project discovery and do not include user-supplied arguments.
[COMMAND_EXECUTION]: The skill requires the Bash tool to perform repository analysis. It uses standard shell commands like find, cat, and grep to inspect files and validate directory structures as part of its auditing procedure.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from the repository being audited (including YAML schemas, Python code, and GraphQL queries). While the skill lacks explicit boundary markers or sanitization for this external content, the risk is inherent to the auditing function and is mitigated by the agent's internal safety guardrails.
Ingestion points: Reads repository configuration, schemas, objects, and Python scripts during the audit phase.
Boundary markers: None identified in the prompt templates.
Capability inventory: Uses Bash, Read, Grep, and Glob to interact with the file system.
Sanitization: No explicit sanitization or escaping of file content is described before processing.

infrahub-auditing-repo