A goal-like loop that decomposes work into systematic, evidence-bound ultrawork steps. Use when the user wants a goal loop or durable, checkpointed execution.
81%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
The canonical home for this skill is ulw-loop in code-yeongyu/oh-my-openagent
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required define-goal step shapes success criteria around reading GitHub PR review-comment threads via `gh pr view`, so the workflow can ingest outsider-authored PR review comments.
code-yeongyu/lazycodex
7e2aae8
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.