Install any skill in seconds. Free to start, no credit card required.
Get Started Free →chorus-aidlc/docs-chorus
Chorus documentation router — consult the live Chorus docs site to answer product-usage questions (UI workflow, agent/plugin setup, API/MCP, deployment, operations).
Passed 7 of 14 test cases with the skill, and 2 of 14 without it.
and 12 more in the full report
8 fixed3 passed either way12 still failing0 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-3.8-flash on 10/18/2026 did not produce a publishable comparison (not comparable). The snapshot above is the previously measured run.
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
I want to claim idea #104 in Chorus and move it into the proposal phase in the AI-DLC pipeline.
| Checked | Without skill | With skill |
|---|---|---|
| The response does not query https://doc.chorus-ai.dev/llms.txt. | Passed | Passed |
| The response indicates that managing pipeline ideas is handled by workflow capabilities. | Missed | Passed |
The instructions your model reads when the skill loads.
This skill is a thin router to the live Chorus documentation site (https://doc.chorus-ai.dev). Use it to answer questions about how to use, configure, deploy, or operate Chorus — grounding the answer in the current published docs instead of memory.
It is not a workflow skill: it does not drive the AI-DLC pipeline. For that, use idea-chorus, proposal-chorus, develop-chorus, review-chorus, or yolo-chorus.
Use this skill whenever the user asks a product-usage question about Chorus, such as:
Do NOT use it to drive the pipeline (claiming ideas, writing proposals, executing tasks) — that is what the stage skills above are for. This skill answers "how does the product work / how do I set it up"; the stage skills do the work.
The docs site is agent-friendly. Follow this three-step convention every time. Do NOT answer from memory, and do NOT hardcode a page list — the index is the source of truth and pages change over time.
https://doc.chorus-ai.dev/llms.txt — a machine-readable index that lists every documentation page with a one-line summary and its .md URL. The index is a single, unlocalized file that lives ONLY at the root /llms.txt. Never prefix it with a locale — https://doc.chorus-ai.dev/zh/llms.txt (and /ja/, /ko/) does NOT exist and returns 404..md to the page URL (e.g. https://doc.chorus-ai.dev/guides/getting-started → https://doc.chorus-ai.dev/guides/getting-started.md)..md URL without the .md suffix) so the user can open it in a browser.Use whatever web-fetch capability your environment provides (your built-in fetch tool, curl, etc.) — this skill states the convention, not a specific tool binding.
The /llms.txt index itself is not localized — there is exactly one, at the root. Localization applies to pages, not the index:
https://doc.chorus-ai.dev/llms.txt and lists the root (en) page URLs. Do not look for /zh/llms.txt — it does not exist.test cases passed, vs 2 / 14 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.