The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to create and execute a shell script (init.sh) to initialize the project environment, including starting services and running migrations.
[EXTERNAL_DOWNLOADS]: The generated initialization script performs package installations using npm install, which fetches code from external registries.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads context from external files and git history without explicit boundary markers or sanitization.
Ingestion points: Files like claude-progress.md, docs/specs/, docs/plans/, and the output of git log are read to reconstruct session state.
Boundary markers: The instructions do not define delimiters or warnings to ignore instructions within these files.
Capability inventory: The agent has the ability to execute shell commands, install packages, and modify the file system.
Sanitization: There is no mechanism to validate or sanitize the content retrieved from project documentation or logs before processing.

task-execute