The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs multiple shell operations including mkdir, cp, rsync, grep, and git rev-parse to set up directory structures and manage files. It also checks for the Python version at runtime.
[INDIRECT_PROMPT_INJECTION]: The skill has a vulnerability surface where it ingests untrusted data that could influence its behavior.
Ingestion points: User-provided inputs for space key, name, about, and owner are collected during the interview step (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content embedded within these user-provided strings before they are written to docs/docs.json or index.md files.
Capability inventory: The skill has the ability to execute python3 scripts (build_docs.py, run.py), write to the filesystem, and modify CI/CD workflows (.github/workflows/).
Sanitization: There is no evidence of sanitization or validation performed on the user inputs before interpolation into configuration files or documentation templates.
[DYNAMIC_EXECUTION]: The skill copies a Python script (build_docs.py) from the plugin's template directory to the local scripts/ folder and subsequently executes it using python3. While the source is internal, this pattern involves executing code that is moved during the skill's lifecycle.
[COMMAND_EXECUTION]: The skill utilizes the $POWERSUIT_OVERLAY environment variable to conditionally execute shell commands and load external configuration files. This allows for runtime behavioral changes based on the local environment state.

docs-init