The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the user/agent to execute several shell commands involving npx tsx to run local scripts such as scripts/plan-lifecycle.ts, scripts/apply-lifecycle.ts, and scripts/run-spam-test.ts. These scripts perform operational tasks like domain management and spam testing.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and analysis of external data (cold email replies) through the /positive-reply-scoring tool.
Ingestion points: External email replies are processed during the "Wednesday — Positive-reply sweep" and "Friday — Campaign retrospectives" phases.
Boundary markers: None are explicitly defined in this playbook to separate untrusted email content from agent instructions.
Capability inventory: The skill has the capability to execute local scripts via npx tsx and write to the local file system (e.g., profiles/<slug>/experiments/*.json and profiles/<slug>/retrospectives/*.md).
Sanitization: There is no mention of sanitization or filtering for the external email content before it is scored or reviewed.
[DYNAMIC_EXECUTION]: The use of npx tsx for runtime execution of TypeScript scripts is a dynamic execution pattern, although it is directed at local project files rather than remote sources.

cold-email-weekly-rhythm