The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard command-line tools including git, rg (ripgrep), and bun to manage repository state and validate code changes. These operations are essential to the skill's primary function and are used transparently.
[DATA_EXPOSURE_AND_EXFILTRATION]: The instructions involve reading repository metadata, commit history, and source code content. This data access is localized to the repository environment and is used solely for auditing the integration process; no network exfiltration patterns were identified.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the Git repository (commit messages and source code), creating a potential surface for indirect injection.
Ingestion points: Data is ingested via git log, git show, and rg commands as seen in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded prompts within the processed Git data.
Capability inventory: The skill can execute shell commands (git) and build scripts (bun) on the local system.
Sanitization: No sanitization or filtering is applied to the repository content before it is processed by the agent.

manual-branch-integrator