The Agent Skills Directory

[COMMAND_EXECUTION]: Executes shell commands to manage vendor-specific software and authentication.
Evidence: Runs npm list -g @opendeploydev/cli, npm view, and opendeploy CLI tools to verify installation and check authentication status (SKILL.md).
[DATA_EXFILTRATION]: Accesses vendor-specific local credential files.
Evidence: Manages authentication by reading and writing to ~/.opendeploy/auth.json, which contains API keys and gateway configurations (SKILL.md).
[PROMPT_INJECTION]: Indirect prompt injection attack surface identified in the data processing flow.
Ingestion points: The agent processes output from external CLI tools (npm, opendeploy) which could be influenced by external environment or API responses (SKILL.md).
Boundary markers: Absent; there are no explicit delimiters or instructions to ignore embedded commands in CLI output.
Capability inventory: The skill has the ability to execute subprocesses via the npm and opendeploy commands (SKILL.md).
Sanitization: No sanitization or validation of CLI output is specified before processing by the agent.

opendeploy-auth