The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the @chandrashekharchoudha/bifrost-cli package globally via npm if it is not already present on the system.
[COMMAND_EXECUTION]: Utilizes multiple shell tools including the native bifrost binary, git, gh, and npm to perform repository management, environment setup, and deployment operations.
[DATA_EXFILTRATION]: Accesses sensitive local files such as .env, .env.local, and .bifrost.yaml to retrieve configuration and credentials required for platform authentication and application deployment.
[PROMPT_INJECTION]: Contains a vulnerability surface for indirect prompt injection because the skill instructs the agent to read and interpret data from repository-level files that could be controlled by an attacker.
Ingestion points: Processes files such as package.json, Dockerfile, next.config.js, and .env from the local repository during preflight and deployment phases.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when the agent reads these external files.
Capability inventory: The skill has access to shell execution (bifrost, git, gh, npm) and file system write capabilities.
Sanitization: There is no evidence of sanitization or strict schema validation performed on the contents of the ingested repository files.

bifrost-platform