The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes various Git CLI tools (git remote, git fetch, git pull, git status, git commit, git push) to perform its primary function. These commands interact directly with the underlying operating system's shell.
[EXTERNAL_DOWNLOADS]: The skill performs network synchronization by fetching and pulling data from configured Git remotes. This involves downloading data from external sources specified in the repository's configuration.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks due to its data processing workflow.
Ingestion points: The agent reads file names from git status, analyzes the content of changed documents to summarize them, and parses the wiki-manifest.yaml file for conflict resolution.
Boundary markers: There are no explicit markers or instructions to isolate the content of documents from the agent's instructions during the summarization phase.
Capability inventory: The skill possesses the ability to perform file system modifications (via Git) and network transmissions (git push).
Sanitization: The skill does not describe any sanitization logic for the content it reads. However, the workflow enforces a human-in-the-loop checkpoint using the AskUserQuestion tool before any changes are committed or pushed, which significantly reduces the risk of automated exploitation.

wiki-commit