The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructions direct the agent to read sensitive configuration files, specifically ~/.hermes/.env to check the API_SERVER_KEY and ~/.cc-connect/config.toml to verify management and bridge tokens.
[COMMAND_EXECUTION]: The skill manages system service persistence by executing commands to stop, uninstall, and install daemons (cc-connect daemon install). It also uses lsof to monitor network ports.
[EXTERNAL_DOWNLOADS]: The skill performs global installation of software using npm install -g cc-connect@latest and references an external configuration guide on GitHub (https://raw.githubusercontent.com/chenhg5/cc-connect/refs/heads/main/INSTALL.md).
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes external data.
Ingestion points: The skill reads external data from cc-connect daemon logs and an external INSTALL.md guide.
Boundary markers: No delimiters or instructions to ignore embedded commands are present when processing this data.
Capability inventory: The skill has the capability to execute shell commands, manage system daemons, and install packages via npm.
Sanitization: No sanitization or validation of the ingested data is performed before it is processed by the agent.

clawpilot-doctor