The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute curl for interacting with vendor APIs and python3 for parsing JSON response data.
[CREDENTIALS_UNSAFE]: The skill manages sensitive API keys for Jetty and AI providers (OpenAI/Gemini).
It stores the Jetty API token in a local CLAUDE.md file for use in subsequent sessions.
It transmits provider API keys to Jetty's servers at dock.jetty.io to store them in collection environment variables.
[EXTERNAL_DOWNLOADS]: The skill fetches a generated image from the flows-api.jetty.io API and saves it to the local project directory.
[PROMPT_INJECTION]: The skill processes data from external sources that could potentially contain malicious instructions.
Ingestion points: Reads tokens from the local CLAUDE.md file and status data from the Jetty API.
Boundary markers: No specific delimiters or "ignore previous instruction" warnings are used when interpolating these values.
Capability inventory: The skill possesses Bash (network and command execution) and Write (filesystem access) capabilities.
Sanitization: The skill employs Python's json module to parse structured data from API responses, which helps prevent direct execution of malicious payloads embedded in strings.

jetty-setup