The Agent Skills Directory

[SAFE]: The skill performs project management tasks such as reading project documentation and interacting with an issue tracker to automate task creation.
[SAFE]: A mandatory human-in-the-loop (HITL) step is included in the process. The agent must present the proposed issue breakdown to the user for approval before any issues are published to the project's tracker. This process effectively mitigates risks associated with processing untrusted external data, such as Indirect Prompt Injection, as the user reviews the generated content before action is taken.
[SAFE]: No evidence of obfuscation, hardcoded credentials, unauthorized network access, or malicious command execution was found in the skill instructions.

to-issues