The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to instructions embedded in the data it processes during the conflict resolution workflow.
Ingestion points: The skill reads local file contents to locate and interpret git conflict markers (<<<<<<<, =======, >>>>>>>) in SKILL.md under the 'Action: resolve' section.
Boundary markers: There are no explicit boundary markers or instructions to the agent to ignore natural language instructions that might be contained within the source code or text being merged.
Capability inventory: The skill has the capability to write to the local file system (to resolve conflicts) and execute git commands such as git add, git commit, and git rebase --continue in SKILL.md.
Sanitization: The skill lacks sanitization or validation of the content read from files before it is processed by the agent to understand the 'intent' of the code changes.

git-merge