The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the bru CLI to import API specifications into local collections. This execution is managed by instructions to check for local installation first.
[REMOTE_CODE_EXECUTION]: The instructions facilitate the use of npx to run the @usebruno/cli package. Security is addressed by requiring explicit user approval, version pinning, and clear communication of risks to the user.
[EXTERNAL_DOWNLOADS]: To analyze application routes, the skill performs shallow clones of remote Git repositories into a temporary session environment.
[DATA_EXFILTRATION]: While the skill reads application source code and specifications, it incorporates specific rules to replace real secrets, tokens, and environment variables with placeholders, preventing accidental exposure in the generated output.
[PROMPT_INJECTION]: The skill processes untrusted content from remote repositories and API contracts, presenting an indirect prompt injection surface.
Ingestion points: Remote Git repositories, OpenAPI/WSDL spec files, and local application code (processed in SKILL.md).
Boundary markers: No explicit delimiters are specified for separating processed data from agent instructions.
Capability inventory: Shell execution via bru/npx and file system writes (SKILL.md, BRUNO.md).
Sanitization: Implements filename sanitization and secret-removal logic, though it lacks general data sanitization for the ingested source content.

api-to-bruno