The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to execute standard git and gh (GitHub CLI) commands to fetch updates, rebase branches, and reset branch tips. These operations are within the expected scope for the skill's stated purpose of repository synchronization and include safety measures such as the use of --force-with-lease for push operations.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as GitHub PR metadata and remote repository states, creating a vulnerability surface for indirect prompt injection.
Ingestion points: The agent is instructed to read external content using git fetch, gh pr view, and gh repo view to determine mergeability and ancestry.
Boundary markers: There are no explicit delimiters defined to separate the output of these external tools from the agent's instructions, which could allow maliciously crafted repository data to influence agent behavior.
Capability inventory: The skill grants the agent the ability to perform high-impact actions including git rebase, git reset --hard, and git push, which could be exploited if the agent follows instructions embedded in untrusted repository metadata.
Sanitization: The skill does not implement validation or sanitization of strings returned by the GitHub API or git log before they are processed by the agent.

fork-upstream-sync