The Agent Skills Directory

[PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it is designed to process untrusted data from external sources provided by the user.
Ingestion points: The skill accepts git log outputs, task lists from Jira/Feishu, and meeting notes (SKILL.md, Step 1).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard embedded commands within the processed data.
Capability inventory: The skill is limited to text processing and does not have access to dangerous capabilities like shell execution, file system modification, or network requests.
Sanitization: There is no evidence of input validation or sanitization of the user-provided logs before they are processed.
[EXTERNAL_DOWNLOADS]: The skill contains a recommendation for the user to install an additional resource from a GitHub repository (mindverse/skillhub). This is presented as an optional manual action for the user to enhance their experience and does not involve automated background downloads or execution.

weekly-report