Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →tldraw/take
Find a GitHub issue in tldraw/tldraw, assign it, implement it, verify it, and open a pull request.
Passed 9 of 16 test cases with the skill, and 8 of 16 without it.
and 2 more in the full report
5 fixed8 passed either way6 still failing4 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
A newer attempt on gemini-3.6-flash on 10/2/2026 did not produce a publishable comparison (not comparable). The snapshot above is the previously measured run.
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
If a tldraw issue description lacks technical detail and technical notes, what codebase step should be performed before determining whether implementation is safe?
| Checked | Without skill | With skill |
|---|---|---|
| Explores the codebase to assess implementation safety before deciding on next steps. | Missed | Passed |
The instructions your model reads when the skill loads.
Find an issue in tldraw/tldraw, implement it, and open a pull request.
The user may reference an issue by number, URL, or description.
gh issue view 123 --repo tldraw/tldraw
gh issue list --repo tldraw/tldraw --search "dark mode" --state open --limit 10
gh issue list --repo tldraw/tldraw --search "dark mode" --state all --limit 10
If there is one clear match, proceed. If several issues match, ask the user to choose from issue numbers and titles. If none match, ask whether to create a new issue using the issue skill.
Read the full issue and comments. Identify:
Agent-drafted issues may include ## Confidence and ## Open questions sections. Treat any unanswered Critical: question, _Awaiting answer._ entry, or More Info Needed label as a blocker unless codebase exploration proves the intended behavior is unambiguous. Resolve blockers with the user before implementing; non-critical questions marked _Deferred by user; not blocking implementation._ may remain open.
test cases passed, vs 8 / 16 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.