The Agent Skills Directory

[PROMPT_INJECTION] (SAFE): No attempts to override agent instructions or bypass safety filters were detected. The instructional language is standard for technical documentation.
[DATA_EXFILTRATION] (SAFE): The skill does not access sensitive system files (e.g., SSH keys, AWS credentials). Network activity is restricted to standard version control operations (git push, but push) targeting the user's configured remote repositories.
[COMMAND_EXECUTION] (SAFE): Commands such as but, git, gh, and jq are used for their intended purposes of branch management and pull request creation. There is no evidence of arbitrary or malicious command execution.
[REMOTE_CODE_EXECUTION] (SAFE): No external scripts are downloaded or executed. The skill relies entirely on locally installed CLI tools.
[OBFUSCATION] (SAFE): All files consist of clear, human-readable text and standard shell commands. No hidden characters or encoded payloads were found.

gitbutler-stacks