The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs shell-level operations to manage the project repository.
Evidence: The workflow includes explicit steps to perform git commit and git tag commands based on derived version information.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external files which serves as input for its automated tasks.
Ingestion points: The skill reads data from TASKS.md, AGENTS.md, and individual task documentation files to determine version impact and generate summaries.
Boundary markers: There are no specified delimiters or instructions provided to the agent to distinguish between task data and potentially malicious embedded instructions.
Capability inventory: The agent has permissions to modify local files (CHANGELOG.md, TASKS.md), execute git commands, and interact with external repository hosting APIs.
Sanitization: The workflow does not include steps to sanitize or escape task descriptions before they are included in changelogs or used in command arguments.

release