From wicked-garden
Pulls deeper context from brain and bus when uncertain, to verify assumptions and check prior decisions before acting.
How this skill is triggered — by the user, by Claude, or both
Slash command
/wicked-garden:groundThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are uncertain. Pull what's known into focus.
You are uncertain. Pull what's known into focus.
question argument (free text from the user / Claude's internal state)wicked-brain:query — conceptual grounding ("what do we know about X")wicked-brain:search — specific decisions, patterns, gotchas (top 5 results)wicked-bus:query — recent bus events matching the question (last 50, filter by
relevance to question terms)brain/memory, brain/wiki, brain/chunk, or bus/eventwicked-brain:read {path} depth=2){most relevant path}, use
wicked-brain:read {path} depth=2"When invoked with a question:
Step 1 — Decompose the question into 3–5 search terms. Extract noun phrases,
named entities, and technical terms. Example: "v8 daemon projection model" →
["daemon", "projection", "v8 architecture", "state machine"].
Step 2 — Parallel execution. Invoke all three in a single parallel batch:
# Brain conceptual query
Skill(wicked-brain:query, question="{question}", session_id="{session}")
# Brain symbol/decision search (repeat per term if ≥2 terms)
Skill(wicked-brain:search, query="{term1}", limit=5, session_id="{session}")
Skill(wicked-brain:search, query="{term2}", limit=5, session_id="{session}")
# Bus recent events
Skill(wicked-bus:query, query="{question}", limit=50)
Step 3 — Rank and dedupe. Collect all results. Score by:
Keep the top 5–10 unique signals. Drop results where two sources say the same thing — keep the higher-priority source.
Step 4 — Format output. Use this shape:
## Grounding: {question}
### What the brain knows
1. [brain/memory] {one-line relevance} — `{path}` → suggest: wicked-brain:read {path}
2. [brain/wiki] {one-line relevance} — `{path}`
3. [brain/chunk] {one-line relevance} — `{path}`
### Recent bus activity
4. [bus/event] {event_type} @ {timestamp} — {one-line relevance}
5. [bus/event] {event_type} @ {timestamp} — {one-line relevance}
### If you need more depth
`wicked-brain:read {most relevant path} depth=2`
Step 5 — If zero results from both brain and bus, say so explicitly:
"No prior decisions or recent events found for this question. Proceeding without
grounding — consider storing the decision you reach with wicked-brain:memory."
Never block progress. Ground is a focusing tool — absence of prior context is itself a useful signal.
If you reach a decision that others should know about:
wicked-brain:memory (store mode)wicked-bus:emit with the relevant event typeThe value of grounding compounds when decisions are written back.
Provides 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.
npx claudepluginhub mikeparcewski/wicked-garden --plugin wicked-garden