The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute build commands extracted from the repository's continuous integration (CI) configuration files. While necessary for building the application before testing, this involves the execution of arbitrary scripts defined within the codebase.
[INDIRECT_PROMPT_INJECTION]: The skill processes instructions from external project files to determine its execution steps, creating a potential vulnerability surface.
Ingestion points: The skill reads configuration from files such as .github/workflows/*.yml, .gitlab-ci.yml, and bitbucket-pipelines.yml to identify build steps.
Boundary markers: There are no specific delimiters or safety warnings provided to the agent to help it distinguish between legitimate build instructions and potentially malicious commands embedded in these files.
Capability inventory: The agent uses shell command execution to run build scripts and the meticulous CLI for uploading assets and triggering remote test runs.
Sanitization: The skill lacks mechanisms to validate or sanitize the commands found in the repository's configuration files before they are executed in the local environment.

meticulous-test