The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local project binary located at ./bin/hushh. This tool is used for retrieving protocol help information, launching local development terminal environments (e.g., backend and stack modes), and managing backend operations.
[COMMAND_EXECUTION]: The skill invokes pytest via python3 -m pytest to execute backend tests within the consent-protocol directory as part of its required verification workflow.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests various repository files while maintaining command execution capabilities.
Ingestion points: The skill reads consent-protocol/README.md, consent-protocol/docs/README.md, docs/reference/architecture/architecture.md, and docs/project_context_map.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the documentation ingestion process.
Capability inventory: The skill has the capability to execute shell commands through the ./bin/hushh binary and the pytest testing framework.
Sanitization: There is no evidence of sanitization or validation of the content within the ingested markdown files before they are processed by the agent.

backend