/weekly-performance-advisor

name: weekly-performance-advisor
category: get-qualified-meetings
type: use-case
tags: [analysis]
description: "Generate your Weekly Performance Advisor dashboard from your La Growth Machine data: a two-tab cockpit (To do + Weekly performance) that flags campaigns to fix, sorts replies by urgency, and tracks reply volume week over week. It pulls only YOUR live LGM data, scores each running campaign against baked 3-zone benchmarks, classifies your untagged replies, and renders a live artifact you can re-open every Monday. Use when the user wants a weekly outbound performance dashboard, a Monday cockpit, campaign health at a glance, \"how are my campaigns doing this week\", \"who do I need to reply to\", or a reply-triage + campaign-health view. Triggers on: 'weekly performance dashboard', 'my outbound cockpit', 'weekly performance advisor', 'campaign health this week', 'how are my campaigns doing this week', 'my Monday cockpit'. First run does a short setup (detect + connect LGM, pick the identity, optional deal layer); every run after that just rebuilds the dashboard. Maintained by La Growth Machine."

Weekly Performance Advisor

Generate the user's **own** Weekly Performance Advisor: a two-tab live artifact built from **their** La Growth Machine (LGM) data. This skill is **fully self-contained** — every rule needed to detect the environment, score campaigns, classify replies, and fill the dashboard is written below. It references no other document.

**Two hard rules, always:**

1. **Only ever show the downloading user's data.** Pull exclusively from *their* connected MCPs. Never inject numbers, campaign names, replies, or examples from anyone else. The template ships **empty**; it is filled at runtime with their live values. If you cannot pull a value, render the documented empty state — never a placeholder number and never a value borrowed from somewhere else. 2. **The output is a LIVE ARTIFACT, not a saved HTML file.** In Generate mode you fill the template in memory and render it as a live artifact **using the available artifact tool** (whatever the current Cowork build exposes for creating/updating an artifact — do not hard-code a tool name). On a refresh, **update the existing artifact** rather than creating a new one. The `assets/dashboard-template.html` file is the **build source only**.

---

Two modes

• **Setup mode** (first run, or when LGM isn't connected / no identity chosen yet): interactive.

Detect the environment, guide install if needed, ask the scoping questions. Conversational.

• **Generate mode** (every run once setup is done): run the engine, fill the template, emit the

artifact + a short handoff. Apply output discipline here — return the artifact and a few lines of handoff, **no step-by-step narration**.

If setup was already done in this project (a `./.lgm-wpa/` snapshot exists and LGM is connected), skip straight to Generate mode; only re-ask a question if something is missing.

---

Phase 0 — Detect (silent, no questions)

Check your **own tool list** — do not ask the user what they have installed.

• **LGM MCP present?** True if tools named `mcp__*` for La Growth Machine are available, i.e. any of

`list_identities`, `list_campaigns`, `get_campaign_stats`, `get_conversations_to_reply`, `get_conversation_messages`, `search_conversations`. (In a terminal you could also confirm with `claude mcp list` showing an LGM server, but tool presence is enough — never ask the user.)

• **CRM / deal MCP present?** Is a HubSpot MCP available (deal/company/contact tools)?
• **`campaign-impact-analyzer` skill available?** (used to build the optional deal/€ layer.)

Record these three booleans and move on. Do not announce the detection.

---

Phase 1 — LGM MCP gate

**If the LGM MCP is present:** say "LGM connected ✓" in one line and go to Phase 2.

**If the LGM MCP is absent:** the dashboard is built entirely from LGM data, so it can't run yet. Explain that in one sentence, then guide the install (do not ask "is it installed?" — you already know it isn't):

1. From the `gtm-system` repo, run the installer and complete browser sign-in:

   sh install.sh

(or the curl one-liner the repo's README provides). Auth is **browser OAuth** — there is **no API key to paste**. 2. No LGM account yet → point them to register at https://app.lagrowthmachine.com and then re-run the installer. 3. After they've run it, **re-detect** (Phase 0). Loop until the LGM tools appear, then continue.

Keep this friendly and short; it's a gate, not a lecture.

---

Phase 2 — Identity scope (a real question)

Call `list_identities`. Then, to pick a sensible default, get each identity's campaign load: call `list_campaigns {status:"RUNNING"}` (paginate) and count RUNNING campaigns per `identity.id`.

Ask which identity to track, **defaulting to the identity running the most RUNNING campaigns** (name it, and ask them to confirm or switch). Mention they can add more identities later for a team-wide view (the dashboard supports multiple identities — union their campaigns and replies).

Store the selected identity id(s) and display name(s); every LGM pull below is scoped to them.

---

Phase 3 — Deal layer (optional, additive — never blocks)

The dashboard ships a **"Deals from campaigns"** box. Without a CRM it shows `—` and the note *"connect HubSpot to pull deal data"*. Offer the richer layer, but **never block the build on it.**

• **CRM/deal MCP present** → offer now: "Want me to add the deals/€ layer (which campaigns actually

drove pipeline), or ship the dashboard first and add it later?"

• **No CRM MCP** → ask "What do you track deals in?" (HubSpot / Salesforce / Sheets / other / none).
• HubSpot but not connected → offer to connect it (then it becomes "present").
• Salesforce / Sheets / other / complex → **dashboard-first**; add the deal layer later.

**Delegation (do not re-implement the join here):**

• If `campaign-im