The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes various subcommands of the basecamp CLI, including doctor, auth login, and setup. These operations are used to diagnose system state and perform automated configuration within the scope of the Basecamp environment.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and interpret structured JSON output from the basecamp doctor command. It uses data from the status, hint, and breadcrumbs fields to decide on remediation steps.
Ingestion points: Data is received from the output of the basecamp doctor --json command execution as described in SKILL.md.
Boundary markers: No explicit delimiters are specified for the tool output, though it is expected in a structured format.
Capability inventory: The skill has the ability to execute file-system impacting commands via the basecamp setup suite and authentication flows via basecamp auth login across SKILL.md.
Sanitization: The instructions tell the agent to look for specific key-value pairs (pass/warn/skip/fail) to guide its logic, which acts as a form of implicit validation.
[SAFE]: The instructions include a specific security constraint that prevents the agent from reading, printing, or requesting credential files, which reduces the risk of accidental data exposure.

basecamp-doctor