The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to interact with the file system and external tools. It executes git for repository status, history analysis, and commit/push operations, and dotnet build for project compilation. These commands are executed within a local environment defined by hardcoded paths.
[DATA_EXFILTRATION]: The skill contains hardcoded absolute Windows file paths (e.g., C:\Users\liveu\source\repos\...) which exposes the local system username (liveu). This constitutes PII (Personally Identifiable Information) exposure if the skill is shared.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from an external upstream repository by reading git commit messages, logs, and code diffs.
Ingestion points: Reads git logs and diffs via git log and git diff from the upstream ShareX repository.
Boundary markers: The instructions lack explicit boundary markers or delimiters to prevent the agent from following instructions embedded within commit messages or code comments.
Capability inventory: The skill has capabilities to write files to the local file system, execute build commands (dotnet build), and push changes back to repositories (git push).
Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the repository history before it is processed by the agent.

port-imageeditor