From growthub-governed-console
Drafts dashboards, widgets, API registry rows, custom objects, and swarm graphs for a governed Growthub workspace via a propose-only planning engine with explicit apply step.
How this skill is triggered — by the user, by Claude, or both
Slash command
/growthub-governed-console:workspace-helperThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
The workspace helper is a governed, workspace-grammar-aware planning engine.
The workspace helper is a governed, workspace-grammar-aware planning engine.
It never mutates on query: query returns proposals; a separate apply
step validates and writes them through the same policy gate as every other
mutation, and every apply appends a receipt to source-records.
# Query — no writes, returns proposals[]
growthub workspace helper query --intent <intent> --prompt "<brief>" --json > proposals.json
# Apply — validates + writes accepted proposals (explicit human/agent review between)
growthub workspace helper apply --proposal-file proposals.json [--yes]
# Receipt history
growthub workspace helper receipts [--limit 25]
Intents: build_dashboard | create_widget | register_api |
create_object | edit_view | repair | explain | swarm
POST /api/workspace/helper/query → { summary, proposals[], warnings[], receipts } (no writes)POST /api/workspace/helper/apply → { applied[], skipped[], workspaceConfig }GET /api/workspace/helper/receipts → apply receipt historyquery with the narrowest intent that matches the ask.preflight_patch / find_downstream_dependencies (this plugin's
governed-console skill) before applying.apply only accepted proposals. Never edit the proposal file to smuggle
in changes the helper did not draft — hand-built mutations go through the
normal PATCH lane instead.applied[] and the receipts stream.dashboards, widgetTypes, canvas, dataModel)
is the hard ceiling — helper applies cannot exceed it.helper/apply is operator-privileged under app scope — expect a structured
scope violation (with repairPlan[]) if scoped incorrectly, and follow it.query lane is meaningful;
do not fabricate an apply result.npx claudepluginhub growthub-ai/growthub-local --plugin growthub-governed-consoleOperates governed-universe MCP tools over a Growthub workspace: describe, traverse metadata graph, simulate causal impact and blast radius, check ship-readiness, dry-run patches, and convert findings into governed next actions.
Set up an agentic OS as a standalone Next.js web dashboard — wired to the Claude Agent SDK plus up to 7 live MCP integrations (Circle community, Fireflies meetings, YouTube/VidIQ, Unipile LinkedIn DMs, Apify Twitter/X, Reddit), per-profile views, a button-bar of actions, snapshot-style data refreshes, and an optional Railway deploy with HTTP basic auth and a persistent volume. A real web app — runs locally or live. Use when the user says "set up a standalone agentic OS", "build an MCP-powered web dashboard", "spin up my AI ops dashboard", "deploy an AI dashboard to Railway", "give me a web version of the dashboard", or asks to personalize a web dashboard previously created with this skill.
Builds or updates PostHog dashboards from insights, reusing existing insights or creating new ones. Uses PostHog dashboard templates as reference for metric groupings.