An always-on unknown-unknowns system for Claude Code. Scout converts unknown unknowns into known unknowns before you commit to an approach: the terms of art, taxonomies, canonical tools, named methods, and tribal knowledge of any domain — plus the questions you didn't know to ask.
The story of how it was designed: https://scout-site-gules.vercel.app (that's index.html in this repo)
curl -fsSL https://raw.githubusercontent.com/tripleyak/scout-site/master/install.sh | bashThen restart Claude Code (skills load at session start) and try:
/scout "web animation and motion design"
Requirements: Claude Code, python3, git.
| Piece | Where | Role |
|---|---|---|
/scout skill (SKILL.md + METHOD.md) |
~/.claude/skills/scout/ |
The brain — scoping, blind-lens sweeps, brief compilation, routing |
| Radar hook | ~/.claude/hooks/scout_radar.py + registered in settings.json |
Always-on regex tripwire on every prompt (<50ms, 1/session, 3/day caps) |
| Grading rubric | ~/.claude/rubrics/scout-brief.md |
Briefs are graded by a fresh-context verifier — never self-certified |
| Verifier agent | ~/.claude/agents/verifier.md |
Independent rubric grader (installed only if you don't already have one) |
| Knowledge base | ~/kb/scout/ (raw/, wiki/, outputs/ + compile protocol) |
One brief per domain; re-scouts are delta passes, never duplicates |
| State | ~/.claude/state/scout/ |
Ledger of scouted domains + radar rate-limit state (created empty) |
The installer is idempotent and backs up any existing ~/.claude/skills/scout/ before replacing it. It never overwrites knowledge-base content or your existing verifier agent.
- Radar — a deliberately dumb regex hook watches every prompt for uncertainty phrasing ("I don't even know what's possible") and new-endeavor signals. Cheap trigger, smart filter: the session model decides whether to surface anything.
- Sweep —
/scout <topic>spawns parallel research agents that are blind to each other, each owning one lens: Lexicon & Map, Toolsmith, Elder (tribal knowledge), plus a Territory lens that runs a blindspot pass over your local repo when the task touches a codebase. Diversity catches what redundancy can't. - Interview — the sweep only covers what the web can know. Scout then asks you 3–5 prioritized questions (architecture-altering first) to surface your unknown knowns — the tacit context no search can find.
- Brief — fifteen sections filed to
~/kb/scout/wiki/: a premise check (STRONG/QUESTIONABLE/WEAK verdict on whether the territory is worth mapping at all), vocabulary (≥15 terms), territory map, canonical resources, tribal knowledge, local territory (file:line), an options landscape with S–XL effort tags and a stated recommendation (flip condition + biggest risk — never a bare "it depends"), the questions you didn't know to ask, a four-quadrant gap diff, a watch-list of unknowns likely to surface mid-build, and a ready-to-paste prompt scaffold for whatever prompt or plan comes next. - Verify — a fresh-context verifier grades the brief against the rubric. Required failures send it back.
- Compound — the ledger grows; the radar reads it and gets quieter as your map gets larger. After you build,
/scout debrief <slug>folds what the territory taught you back into the brief.
Modes: /scout quick <topic> (inline, nothing filed) · /scout <topic> (standard) · /scout deep <topic> (6 lenses + completeness critic) · /scout debrief <slug> (post-build delta).
Scout's blind-lens sweep is original; the four-quadrant unknowns frame (known/unknown × known/unknown), the local blindspot pass, the user interview, the prompt scaffold with its specificity-balance rule, and the debrief loop are adopted from Thariq Shihipar's "A Field Guide to Fable: Finding Your Unknowns" (July 2026). His thesis — the map is not the territory, and output quality is bounded by how well you discover your own unknowns before prompting — is exactly the job scout automates.
The cheap-model sweep lanes ("spend the expensive model on judgment, cheap models on volume") and the five-part spec shape for delegated-build scaffolds are adapted from DannyMac180/fable-advisor's architect pattern.
The premise check (assumption / inaction cost / 80% alternative / hard verdict), the no-"it-depends" recommendation rule (stated preference + flip condition + biggest risk), scope-and-risk effort tags, and the one-batched-message interview discipline are adapted from the brainstorm skill in indigoai-us/hq-core — anti-sycophancy as structure, not tone.
Genome — scout maps territory ahead; genome dissects ideas behind (12-layer idea autopsies → generative principles).
rm -rf ~/.claude/skills/scout ~/.claude/hooks/scout_radar.py ~/.claude/rubrics/scout-brief.md
# then remove the scout_radar.py entry from hooks.UserPromptSubmit in ~/.claude/settings.json
# your briefs in ~/kb/scout/ are yours — delete only if you want toMIT