The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill processes file content that may be influenced by external contributors. 1. Ingestion points: Content is read via 'cat' and 'git diff' commands. 2. Boundary markers: No explicit markers are used to isolate conflicting code from instructions. 3. Capability inventory: The skill uses 'Bash', 'Read', 'Edit', and 'git' tools, allowing it to modify files and run tests. 4. Sanitization: No sanitization is performed on the code content. This risk is inherent to the primary purpose of conflict resolution.
[Unverifiable Dependencies] (SAFE): Mentions standard package manager commands like 'pnpm install' and 'npx' for verifying fixes, which is appropriate for the skill's use case.

conflict-resolution