The Agent Skills Directory

[SAFE]: The skill's primary function is to transform text-based audit data into structured work items. All operations described are logical transformations of natural language content.
[COMMAND_EXECUTION]: The skill uses restricted shell commands like cat and ls via the Bash tool to read files and verify local path integrity, which is appropriate for its stated purpose of parsing audit reports.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted data from external audit reports, it treats this content as data for parsing and classification. It does not perform high-risk actions like network requests or dynamic code execution based on the ingested content, effectively mitigating the risk of indirect prompt injection.
[DATA_EXFILTRATION]: The skill explicitly states that exports to trackers (Jira, Linear, etc.) should be generated as local CSV or JSON files and specifies that no API calls should be made, ensuring data remains within the local environment.

backlog-generator