The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes high-impact repository management operations using git and the GitHub CLI.
gh repo edit --visibility public --accept-visibility-change-consequences: Changes repository status from private to public.
git push --force-with-lease origin --all: Overwrites remote repository history during the history rewrite stage.
These commands are core to the skill's purpose and are protected by multiple hard-gated confirmation steps and fresh visibility checks before execution.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted project files to generate public repository metadata and documentation.
Ingestion points: Reads package.json, pyproject.toml, Cargo.toml, and source code entry points (e.g., main modules) to extract project descriptions and topics.
Boundary markers: No specific delimiters or "ignore instructions" warnings are applied to the manifest data being processed.
Capability inventory: The extracted project data is used to construct shell command arguments for gh repo edit --description ... and to write the project's primary README.md file.
Sanitization: The instructions do not define sanitization or escaping protocols for content pulled from manifest files, creating a surface where malicious project metadata could influence the agent's behavior or repository configuration.

oss-launch