docslisted
Install: claude install-skill connorpham/vteam-harness
> **Model routing for this tool** (from `model-routing.data.yaml`, snapshot 2026-08-17):
> `frontier` → **fable** · `workhorse` → **opus** · `standard` → **sonnet** · `utility` → **haiku**
> Roles → tiers: ba-challenger: standard · ba-draft: standard · dev-r1: workhorse · dev-r2: standard (high-stakes: workhorse) · dev-r3: workhorse · explore: utility · qa-challenger: standard · sa-background: workhorse · sa-challenger: workhorse · sa-writer: workhorse
> Resolve at runtime: `python3 .vteam/scripts/model_route.py <role> --tool claude-code [--high-stakes]` — high-stakes diffs (review.high_stakes_*) bump dev-r2 to the workhorse tier.
> Spawning a subagent: pass the resolved name as the Agent tool's `model` parameter.
# /docs — create the documented truth the gates assume (docs lane, no code)
**Why this exists:** every other vteam lane reads an oracle. /ba shards a spec,
/dev cites `docs/specs/<feature>.md`, /qa derives expected values from it, and
the review gate escalates on `review.high_stakes_terms`. A mature repo that grew
without documentation has none of that — so the gates either idle or fire on
nothing. This workflow CREATES that oracle from the only two sources that
actually hold it: **the codebase** and **the owner's head**. It does not require
documentation to pre-exist; it manufactures it, then says out loud which parts
are still guesses.
**What this workflow is NOT:** it never writes or edits product code, never
touches `vteam.config.yaml` (it PROPOSES a patch,