The Agent Skills Directory

[DATA_EXFILTRATION]: The skill facilitates the transmission of project source code to the vendor's platform.
For projects without a git remote, the skill creates a tar.gz archive of the local project directory and uploads it to ploi.cloud using a signed URL and curl.
The archiving process respects .gitignore or .dockerignore files to prevent unintended file exposure.
[COMMAND_EXECUTION]: The skill uses shell commands to automate development workflows.
It executes git for repository discovery and branch identification.
It uses tar for project archiving and curl for network uploads to the Ploi Cloud API.
It performs local file system operations using Read, Glob, and Grep to detect project frameworks and requirements.
[PROMPT_INJECTION]: The skill processes untrusted local files which creates a surface for indirect prompt injection.
Ingestion points: The agent reads and parses configuration files including composer.json, package.json, wp-config.php, and .env.example in the project root.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore embedded instructions within these data files.
Capability inventory: The agent has access to a wide range of powerful tools, including a Bash shell and numerous MCP tools for creating and destroying cloud infrastructure.
Sanitization: No explicit sanitization or validation of the file content is performed before the agent uses the extracted data to configure deployment parameters.

pc-deploy