The Agent Skills Directory

[SAFE]: The skill's instructions focus on legitimate software engineering tasks such as implementing UI changes, refactoring code, and performing technical validation using existing project tools.
[SAFE]: No evidence of prompt injection, data exfiltration, or obfuscation was found. The instructions prioritize following existing repository patterns and pinned dependency versions.
[SAFE]: Although the skill processes external data (repository files and specifications), it does so within the expected scope of a software development assistant. While an inherent surface for indirect prompt injection exists in any tool reading external files, no specific exploitable patterns or dangerous autonomous behaviors were identified.
[SAFE]: The mentioned external tools (ops-browser, ops-client, repo-map) are referenced as project-specific utilities for evidence collection and discovery, consistent with a controlled development environment.

implement-frontend