The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external TYPO3 project management systems, which represents a surface for indirect prompt injection.
Ingestion points: The skill reads issue descriptions from forge.typo3.org and merge request details from git.typo3.org using REST APIs.
Boundary markers: There are no specific delimiters or instructional guardrails implemented to prevent the agent from potentially obeying instructions embedded in the processed external descriptions.
Capability inventory: The skill possesses the ability to execute shell commands (git, ddev), write files (commit messages, local checkouts), and perform network operations to official TYPO3 services.
Sanitization: Scripts correctly use jq and json.dumps to sanitize data for API requests, providing protection against basic injection vectors.
[COMMAND_EXECUTION]: The skill automates complex development tasks by executing local commands through provided scripts.
Evidence: scripts/setup-typo3-coredev.sh executes git, ddev, and docker to configure development environments. scripts/validate-commit-message.py uses subprocess.run to inspect git history. These actions are limited to the skill's primary purpose of facilitating core contributions and use hardcoded commands or sanitized input.
[EXTERNAL_DOWNLOADS]: The skill performs downloads from external sources to set up the development environment.
Evidence: scripts/setup-typo3-coredev.sh clones the TYPO3 CMS source code from its official GitHub repository. As this targets a well-known and trusted organization, the activity is considered part of standard, safe operation.

typo3-core-contributions