The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands like mkdir, rsync, and python3 to set up directories, copy templates from the plugin root, and execute build scripts. It also suggests starting a local web server using python3 -m http.server for testing purposes.
[DYNAMIC_EXECUTION]: The skill copies a Python build script (scripts/build_status.py) from its internal templates to the user's repository and subsequently executes it to generate the status page. It also references the execution of repository-specific scripts like .github/actions/gate/run.py.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through two mechanisms:
Data Ingestion: It processes content from services/*.yml and incidents/incidents.json to generate the public status page. While it contains instructions to ensure text is public-safe, there are no explicit boundary markers or sanitization logic defined in the skill instructions to prevent malicious content in these files from influencing agent behavior.
Dynamic Instruction Loading: The skill includes an 'Overlay' mechanism that checks the $POWERSUIT_OVERLAY environment variable and instructs the agent to read and follow instructions from that external path. This allows for the dynamic injection of instructions into the agent's context from a source outside the static skill definition.
Capability Inventory: The skill possesses the capability to execute shell commands, write files to the local system, and modify CI/CD workflow files (GitHub Actions).

status-page