From wicked-garden
LLM-based work-shape classifier that reads the user's prompt, selects the matching archetype (triage, explore, specify, decide, ship, review, incident, build, migrate, modernize), and identifies signals to steer the session.
How this skill is triggered — by the user, by Claude, or both
Slash command
/wicked-garden:classifyThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are classifying a prompt into a v11 work-shape archetype. Your output
You are classifying a prompt into a v11 work-shape archetype. Your output drives downstream archetype routing for the rest of the session (until the user changes scope or invokes this skill again).
The v11 hook auto-classifier is a regex + boolean-signal heuristic. It works for prompts whose vocabulary matches the catalog phrase lists, but it misses paraphrases and underuses signals. You are a better classifier than regex. This skill is the path to use the model's full reasoning on the classification step, then persist the result so the rest of the session benefits without re-running classification on every turn.
.claude-plugin/archetypes.json defines the work-shape archetypes. Read it
once at the start of this skill. Summary table:
| Archetype | Phases | Use when |
|---|---|---|
| triage | classify | prompt is genuinely ambiguous; ask for clarification |
| explore | frame → diverge → converge | open problem space, multiple paths, brainstorm |
| specify | elicit → structure → validate | requirements / acceptance criteria need writing |
| decide | brief → options → score → record | 2+ viable options, need an ADR |
| ship | canary → ramp → full → soak | already-built change being rolled out |
| review | scope → assess → findings → remediate-or-accept | independent assessment of an artifact |
| incident | triage → investigate → mitigate → resolve → followup | live production failure |
| build | plan → implement → test → review | implement a feature or fix (most common) |
| migrate | plan → expand → backfill → cutover → contract | in-place shape change with rollback proof |
| modernize | discover → extract → blueprint → transform → parity → cutover | port a legacy codebase to a new stack (NOT in-place) |
What is the user actually asking for? Restate in one sentence in your own words. If you can't, the prompt is genuinely ambiguous → triage.
Archetypes are NOT mutually exclusive. Pick a SET. Common combinations:
build + migratereview + shipdecide (and
possibly build if they want you to also implement it)Score each match between 0.0 and 1.0. Use these calibration anchors:
If nothing scores ≥ 0.5, return triage only — that's the signal to ask
for clarification before doing work.
Boolean flags that downstream archetypes use to scale rigor. Mark TRUE only when the prompt clearly implies it:
blast_radius_high — change affects production traffic / many users / many systems.novelty_high — pattern not yet in this codebase.state_complexity_high — touches data shape, migrations, persistent state.reversibility_low — undoing is expensive (data migrations, destructive ops).reversibility_medium_or_low — undoing is non-trivial (config changes, breaking APIs).production_impact — production users / systems affected right now.compliance_scope — GDPR / SOC2 / HIPAA / PCI surface.ambiguity_high — multiple plausible reads.spec_ambiguity_high — success criteria are fuzzy.scope_unclear — boundary of work is undefined.multiple_viable_options — 2+ paths with no obvious winner.post_build — change is already implemented; this is about deployment.code_change — implementation work involved.independent_assessment_needed — someone else's work needs review.Default any flag you didn't explicitly mark to FALSE. Do not over-tag.
Intent is coarser than archetype — used by the hook to gate directive emission. One of:
simple-edit — typo, comment, formatting, single-line fix. Hook stays silent.feature — most non-trivial work (default for build/migrate/ship).rigor — high stakes (compliance, security, blast_radius_high).research — exploratory (explore, decide).Emit a JSON object with the four keys above and pipe to the persist script. Use this exact shape — extras get dropped:
echo '{
"intent": "feature",
"archetypes": [
{"name": "build", "score": 0.85, "evidence": ["implement keyword + code_change signal"]},
{"name": "migrate", "score": 0.65, "evidence": ["schema change + state_complexity signal"]}
],
"signals": {
"code_change": true,
"state_complexity_high": true,
"reversibility_low": true
}
}' | sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" \
"${CLAUDE_PLUGIN_ROOT}/scripts/classify/persist.py"
The script normalises and writes to SessionState. Confirm the response
shows "ok": true, then return control.
After persisting, look at the top archetype's playbook
(skills/archetype/refs/{name}.md) and start running it. Do not re-run
classification mid-session unless the user explicitly changes scope.
/wicked-garden:archetype:<name> directly — they
already classified.classified_at set for this session and the
prompt fits the existing classification. Re-classifying on every turn
is exactly the cost we're trying to avoid.npx claudepluginhub mikeparcewski/wicked-garden --plugin wicked-gardenProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.