The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute a local Node.js script (node src/config.js) to manage settings. This command execution is scoped to the skill's own directory and a specific configuration file.
[INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface as it processes configuration data.
Ingestion points: Data is read from ~/.claude-context-optimizer/config.json when the show or get commands are executed via the Bash tool.
Boundary markers: The instructions do not define specific delimiters or warnings for the agent to treat the configuration file's content as untrusted.
Capability inventory: The skill can execute local commands via Bash and has the ability to read and write to the local filesystem.
Sanitization: The skill description specifies that behavior knobs are "range-validated" and that the script handles errors for bad values or unknown keys, which acts as a safeguard against malicious or malformed input.

cco-config