Common Room

View saved version →

---
name: account-research
description: "Research a company using Common Room data. Triggers on 'research [company]', 'tell me about [domain]', 'pull up signals for [account]', 'what's going on with [company]', or any account-level question."
---

# Account Research

Retrieve and synthesize account information from Common Room. Handles four interaction patterns: full overviews, targeted field questions, sparse data situations, and combined MCP data + LLM reasoning.

## Data Source

Use the Common Room MCP server as the primary source. If its tools are unavailable or unauthenticated, tell the user that the Common Room connection is required and stop before producing account conclusions. Ground every account fact in tool output.

## Step 0: Load User Context (Me)

Before researching any account, fetch the `Me` object from Common Room. This provides:
- The user's profile, title, role, and Persona in CR
- The user's segments ("My Segments")

Default all queries to the user's own segments unless the user explicitly asks for a broader view. This keeps results scoped to their territory.

## Step 1: Identify the Interaction Pattern

Determine what the user actually needs before deciding how much data to fetch:

**Pattern 1 — Full Overview:** "Tell me about Datadog" / "Summarize cloudflare.com"
→ Fetch the full field set and produce a structured briefing.

**Pattern 2 — Targeted Question:** "Who owns the Snowflake account?" / "Is acme.io showing buying signals?" / "What's the employee count for notion.so?"
→ Fetch only the relevant field(s). Return a direct, concise answer — do not produce a full brief for a simple question.

**Pattern 3 — Sparse Data:** "Tell me about tiny-startup.io"
→ If Common Room has limited data for an account, say so honestly: "There is limited information available for this account." Never speculate or fill gaps with generic statements.

**Pattern 4 — Combined Reasoning:** Fetch structured MCP data, then layer in LLM analysis — e.g., "Stripe has 8,000 employees and is hiring heavily for AI roles. Based on your ICP of 1k–10k fintech companies, this is a strong fit."

## Step 2: Look Up the Account

Search Common Room for the account by domain or company name. Exact match first; if no result, try partial match and confirm with the user before proceeding.

## Step 3: Fetch the Right Fields

Use the Common Room object catalog to see available field groups and their contents. For full overviews, request all field groups. For targeted questions, request only what's relevant.

**Key field groups to know about:**
- **Scores** — always return as raw values or percentiles, never labels
- **Summary research** — RoomieAI output; often the richest qualitative signal
- **Top contacts** — sorted by score desc; use communityMemberID for full lookups

**Choosing what to fetch:**

| User query type | Fields to request |
|-----------------|------------------|
| Full account overview | All field groups |
| "Who owns this account?" | Company profiles & links, CRM fields |
| "Is this company a good fit?" | Key fields, scores, about |
| "What signals is this account showing?" | Scores, summary research, CRM fields |
| "Who are the top contacts?" | Top contacts |
| "What does RoomieAI say about them?" | Summary research, all research |
| "Find engineers at this account" | Prospects (with title filter) |

## Step 4: Web Search (Sparse Data Only)

Common Room is the primary data source. Do not run web search when CR returns rich data.

When CR data is sparse (Pattern 3 — few fields returned, no activity, no scores), run a targeted web search to fill gaps:
- `"[company name]" news` — scoped to the last 30 days
- Look for: funding rounds, acquisitions, product launches, executive changes, press coverage

If the user explicitly asks for external context or recent news, run web search regardless of data richness.

## Step 5: Apply Reasoning (Pattern 4)

When the user's question invites synthesis — not just data retrieval — layer in analysis:
- Compare account data to known ICP criteria from session context
- Identify fit signals (size, industry, tech stack, hiring patterns)
- Note timing signals (funding, trial status, recent activity spike)
- Frame insights as clearly derived from data, not assumed

When the user's company context is available (see `references/my-company-context.md`), position findings relative to the user's value proposition and ICP.

## Step 6: Produce Output

Only include sections where Common Room returned actual data. Omit sections entirely rather than filling them with guesses.

**Full overview (when data is rich):**

```
## [Company Name] — Account Overview

**Snapshot**
[2–3 sentences: what they do, plan/stage, relationship status]

**Key Details**
[Employee count, industry, location, domain, funding — from key fields]

**CRM & Ownership** [If CRM fields returned]
[Owner, opp stage, ARR]

**Scores** [If scores returned]
[All available scores as raw values or percentiles]

**Signal Highlights** [If activity/signals exist]
[3–5 most important signals with dates]

**Top Contacts** [If contacts returned]
[Name | Title | Score — top 5 sorted by score desc]

**RoomieAI Research** [If summary research is non-null]
[Summary research output; list all available research topic names]

**Recommended Next Steps**
[2–3 specific, signal-backed actions]
```

**Targeted question:** 1–3 sentence direct answer. No full brief needed.

**Sparse data (few fields returned, most sections would be empty):**

```
## [Company Name] — Account Overview (Limited Data)

**Data available:** [List exactly what Common Room returned]

[Present only the returned fields]

**Web Search**
[Findings from web search — or "No significant recent news found"]

**Note:** Common Room has limited data on this account. The account may need enrichment in Common Room.
```

## Quality Standards

- Scores must always be raw values or percentiles — never categorical labels
- For targeted questions, answer precisely and don't over-deliver
- Be explicit when data is missing or stale — don't speculate
- Keep full briefings readable in 2–3 minutes
- **Every fact must trace to a tool call** — don't include data not returned by Common Room

## Reference Files

- **`references/signals-guide.md`** — signal type taxonomy and interpretation guide
- **`references/me-context.md`** — current-user and My Segments scoping
- **`references/my-company-context.md`** — user-company positioning and ICP context

Referenced files: 6

View saved version →

---
name: call-prep
description: "Prepare for a customer or prospect call using Common Room signals. Triggers on 'prep me for my call with [company]', 'prepare for a meeting with [company]', 'what should I know before talking to [company]', or any call preparation request."
---

# Call Prep

Produce a complete, scannable call prep brief by combining account research, contact research, and signal synthesis from Common Room.

## Data Sources

Use the Common Room MCP server as the primary source. If its tools are unavailable or unauthenticated, tell the user that the Common Room connection is required and stop before producing signal-backed prep. Use an available calendar connector only to identify meeting context and attendees.

## Prep Process

### Step 1: Identify the Account and Attendees

Parse what the user has provided:
- **Company name** — required; look up the account in Common Room
- **Attendee names** — optional; if provided, research each one

**Calendar lookup:** If a calendar connector is available, search for upcoming meetings with the named company to automatically surface attendee names, meeting time, and any meeting notes or agenda. Use this to fill gaps the user didn't provide.

If neither attendees nor a calendar match can be found, ask: "Who will be on the call from [Company]? I can research each attendee to make your prep more useful."

### Step 2: Run Account Research

Use the account-research skill process to build a full account snapshot. For call prep, prioritize:
- Recent product signals (what are they doing in the product right now?)
- Open opportunities or renewal timeline
- Any risk signals (declining usage, support tickets, churned seats)
- Key recent events (funding, executive change, new hire)

When reviewing activity history, prioritize Gong and call recording activities — these provide direct context about previous conversations. Do not filter out call recordings by activity origin.

### Step 3: Run Contact Research for Each Attendee

For each external attendee, use the contact-research skill process. For call prep, focus on:
- Role and influence in the buying process
- Their personal activity and engagement history
- Any recent signals that suggest their current mood/priorities
- Spark persona classification if available

### Step 4: Synthesize Talking Points and Objectives

Based on the combined account and contact research:
- Identify the **call objective** (e.g., discovery, demo, expansion conversation, renewal, QBR)
- Generate **3–5 tailored talking points** grounded in specific signal data
- Anticipate **2–3 likely objections or topics** the customer may raise
- Suggest a **recommended outcome** for the call

When the user's company context is available (see `references/my-company-context.md`), tailor talking points to the user's product and value proposition.

### Step 5: Recency Check (Web Search)

After gathering all Common Room data, run a quick recency check to catch anything that happened since the last CR data sync. This is supplementary — CR data drives the prep; web search only adds recency.

**Company news:** Search `"[company name]" news` filtered to the last 14 days. Look for funding announcements, product launches, leadership changes, layoffs, partnerships, or press coverage.

**Attendee presence:** For each external attendee, search `"[full name]" "[company name]"` — look for recent articles, LinkedIn posts, conference talks, podcasts, or published opinions.

If a company news item is significant (e.g., just raised a round, announced a major hire), flag it in Signal Highlights. Otherwise, include findings briefly — don't let web search results overshadow CR signals.

## Output Format

The output adapts to how much data Common Room returned. Only include sections where you have real data. Never fill a section with invented details.

### When data is rich (multiple field groups returned, activity history, scores, signals):

```
## Call Prep: [Company] — [Date/Time if known]

**Meeting Context**
[Attendees, meeting type, and any known agenda]

---

### Company Snapshot
[4–6 bullets: key account status, signals, and recent activity]

---

### Attendee Profiles

**[Attendee Name] — [Title]**
[3–4 bullets: role, recent activity, Spark persona if available, personal hook]

[Repeat for each attendee]

---

### Signal Highlights
[Top 3 signals most relevant to this specific call]

---

### Talking Points
1. [Point tied to a specific signal]
2. [Point tied to a specific signal]
3. [Point tied to a specific signal]

### Likely Topics / Objections to Prepare For
- [Topic or objection + suggested response]
- [Topic or objection + suggested response]

### Recommended Call Outcome
[1–2 sentences: what success looks like for this meeting]
```

### When data is sparse (few fields returned, no activity, null sparkSummary):

```
## Call Prep: [Company] — [Date/Time if known]

**Data available:** [List exactly what Common Room returned — e.g., "Name, title, email, two tags. No activity history, no scores, no Spark data."]

### What I Found
[Only the fields actually returned, presented as-is]

### Web Search Results
[Findings from web search on the company and attendees — or "No significant results"]

### Suggested Next Steps
- I can pull [specific field groups] from Common Room if available
- I can run deeper web searches on [specific topics]
- You may want to check Common Room directly for [what's missing]
```

Do not generate a full call prep brief from sparse data. A short honest output is always better than a long fabricated one.

## Quality Standards

- Ground every talking point in a real signal — no generic filler
- Keep the brief tight — it should be readable in 5 minutes or less
- Flag unknowns explicitly — if attendee research is thin, say so
- Time-box the research — don't over-research at the expense of speed
- **Never invent deal context** — no fabricated proposals, competitor comparisons, pricing, trial terms, or objections not returned by a tool call

## Reference Files

- **`references/call-types-guide.md`** — guidance for different call types (discovery, expansion, renewal, QBR) and how to tailor prep accordingly

Referenced files: 5

View saved version →

---
name: compose-outreach
description: "Generate personalized outreach messages using Common Room signals. Triggers on 'draft outreach to [person]', 'write an email to [name]', 'compose a message for [contact]', or any outreach drafting request."
---

# Compose Outreach

Generate three personalized outreach formats — email, call script, and LinkedIn message — grounded in Common Room signals for a specific company or contact.

## Data Source

Use the Common Room MCP server as the primary source. If its tools are unavailable or unauthenticated, tell the user that the Common Room connection is required and stop before claiming signal-based personalization. Ground every referenced signal in Common Room output or a cited web result.

## Outreach Process

### Step 1: Look Up the Target

Use Common Room MCP tools to find and retrieve data for the target (company and/or specific contact). Pull:
- Recent product activity and engagement signals
- Community activity (posts, questions, reactions)
- 3rd-party intent signals (job postings, news, funding)
- Relationship history (prior contact, meetings, email opens)

If the user specified a person, run contact-level research. If only a company was given, identify the best contact to target based on title, engagement, and role.

### Step 2: Web Search for External Hooks (If CR Signals Are Thin)

If CR returned strong signals (recent activity, engagement, product usage), those should drive personalization — skip web search. If CR signals are thin or the prospect has little CR activity, run a web search for external hooks:

**What to search:**
- `"[company name]" funding OR acquisition OR launch OR announcement` — last 30 days
- `"[contact full name]" "[company name]"` — look for recent articles, interviews, LinkedIn posts, or conference talks

**Prioritize external hooks that are:**
- Very recent (< 2 weeks) — the prospect is likely still thinking about it
- Publicly visible — they know you could have seen it
- Change-signaling — growth, new role, new product, new market

If the user explicitly asks for web search or external hooks, run it regardless of CR signal richness.

### Step 3: Spark Enrichment (If Available)

If Spark is available, run enrichment on the target contact to get persona classification, background, and influence signals. Use this to calibrate tone and message angle.

### Step 4: Identify the Best Hooks

From the signal data, identify the 1–3 strongest personalization hooks. Rank by:
1. **Recency** — happened in the last 7–14 days
2. **Specificity** — a concrete action they took, not a general trend
3. **Relevance** — connects directly to a value your product delivers

Good hooks: posted a question in the community about X, just hired 5 engineers, recently started using [feature], company just raised Series B, trial nearing expiration, champion just changed jobs.

Bad hooks: "I noticed you're a customer" or generic industry trends.

### Step 5: Generate All Three Formats

Use the strongest hooks to write all three formats. Each format has different constraints and conventions — follow the format-specific guidelines in `references/outreach-formats-guide.md`.

Always produce all three, clearly labeled.

When the user's company context is available (see `references/my-company-context.md`), ground the value bridge and pitch in the user's specific product and positioning.

### Step 6: Annotate Your Choices

After the three drafts, include a brief note (2–4 sentences) explaining:
- Which signals were used and why they were chosen
- Any assumptions made (e.g., inferred call objective)
- Alternative angles if the primary hook doesn't land

## Output Format

```
## Outreach for [Name / Company]

### 📧 Email

**Subject:** [Subject line]

[Email body — 3–5 sentences]

---

### 📞 Call Script

**Opening:**
[Opening line — conversational, 1–2 sentences]

**Value Bridge:**
[Why you're calling and why now — 2–3 sentences tied to a signal]

**Ask:**
[Single, low-friction ask — e.g., 15-minute call, specific question]

---

### 💼 LinkedIn Message

[Under 300 characters. Warm, personal, no pitch.]

---

### Signal Notes
[2–4 sentences: which signals were used, why, and any alternative angles]
```

## When Signal Data Is Sparse

If Common Room returns minimal data on the target (e.g., just name, title, tags — no activity, no scores, no Spark):

1. **Do not draft outreach from thin air.** Outreach grounded in fabricated signals is worse than no outreach.
2. **Run web search first** — this becomes your primary personalization source. Look for recent news, LinkedIn posts, conference talks, company announcements.
3. **If web search also returns little**, present what you have honestly and ask the user for context:

```
## Outreach for [Name / Company] — Limited Data

**What I found:**
[Only the real data from CR and web search]

**I don't have enough signal to draft personalized outreach yet.** To write something strong, I'd need:
- Recent activity or engagement signals
- Context you have from prior conversations
- A specific reason for reaching out now

Can you share any of the above?
```

## Quality Standards

- Every message must reference something specific — generic outreach is not acceptable output
- Match tone to context: warm and conversational for inbound/community signals; more formal for cold/executive outreach
- The LinkedIn message must be under 300 characters — no exceptions
- The call script must be speakable naturally — read it aloud mentally to check rhythm
- **Never fabricate signals** — only reference data retrieved from Common Room or web search

## Reference Files

- **`references/outreach-formats-guide.md`** — detailed format rules, examples, and tone guidelines for each channel

Referenced files: 5

View saved version →

---
name: contact-research
description: "Research a specific person using Common Room data. Triggers on 'who is [name]', 'look up [email]', 'research [contact]', 'is [name] a warm lead', or any contact-level question."
---

# Contact Research

Retrieve a comprehensive contact profile from Common Room. Supports lookup by email, social handle, or name + company. Returns enriched data including activity history, Spark, scores, website visits, and CRM fields.

## Data Source

Use the Common Room MCP server as the primary source. If its tools are unavailable or unauthenticated, tell the user that the Common Room connection is required and stop before producing a contact profile. Ground every contact fact in tool output.

## Step 1: Locate the Contact

Common Room supports multiple lookup methods — use whichever the user has provided:

| What the user gives | Lookup method |
|---------------------|--------------|
| Email address | Look up by email (most reliable) |
| LinkedIn, Twitter/X, or GitHub handle | Look up by social handle — specify handle type explicitly |
| Name + company | Identity resolution by name + org domain; present matches if ambiguous |
| Name only | Search by name; if multiple matches, show a brief list and ask the user to confirm |

If no match is found, respond: "Common Room doesn't have a record for this person." Do not speculate or fabricate profile data.

## Step 2: Fetch Contact Fields

Use the Common Room object catalog to see available field groups and their contents. For full profiles, request all groups. For targeted questions, request only what's relevant.

**Key field groups to know about:**
- **Scores** — always return as raw values or percentiles, never labels
- **Recent activity** — use `Contact Initiated` filter (last 60 days) for their actions, not your team's
- **Website visits** — total count + specific pages (last 12 weeks)
- **Spark** — retrieve all Sparks when tracking engagement evolution over time

## Step 3: Run Spark Enrichment (If Available)

If Spark is available, use it. Spark provides:
- Professional background and job history
- Social presence and influence signals
- Persona classification: Champion, Economic Buyer, Technical Evaluator, End User, or Gatekeeper
- Inferred role in the buying process

If Spark is unavailable but real activity data exists (recent actions, website visits, community engagement), infer a persona from those signals. If neither Spark nor activity data is available, classify as Unknown — do not guess a persona from title alone.

Retrieve **all Sparks** (not just the most recent) when the user wants to understand how this contact's engagement has evolved over time.

## Step 4: Assess Account Context

Pull an abbreviated account snapshot for this contact's parent company. Note:
- Open opportunities, expansion signals, or churn risk at the account level
- Whether other contacts at this company are also active
- How this person's engagement compares to their colleagues

## Step 5: Identify Conversation Angles

Based on activity and signals, surface the strongest 2–3 hooks:
- A recent `Contact Initiated` activity (community post, product event, support ticket)
- A specific web page they visited recently — especially if it signals evaluation intent
- A job change, promotion, or company news
- Their Spark persona and what that suggests about communication style
- Their role in a known active deal

## Output Format

Only include sections where data was actually returned. Omit sections with no data rather than filling them with guesses.

**When data is rich:**

```
## [Contact Name] — Profile

**Overview**
[2 sentences: who they are, their role, and relationship status]

**Details**
- Title: [title]
- Company: [company]
- Email: [email]
- LinkedIn: [URL]
- Other profiles: [Twitter/X, GitHub, CRM link if available]

**Scores** [If scores returned]
[All scores as raw values or percentiles]

**Recent Activity** (last 60 days) [If activity returned]
[3–5 bullets with dates]

**Website Visits** (last 12 weeks) [If visit data exists]
[Total visit count + list of pages visited]

**Spark Profile** [If Spark data is non-null]
[Persona type, background summary, influence signals]

**Segments** [If segments returned]
[List of segment names this contact belongs to]

**Account Context**
[1–2 sentences on their company's status]

**Conversation Starters**
[2–3 specific, signal-backed openers]
```

**When data is sparse (e.g., only name, title, email, tags returned; sparkSummary is null):**

```
## [Contact Name] — Profile (Limited Data)

**Data available:** [List exactly what Common Room returned]

[Present only the returned fields]

**Web Search**
[Any findings from searching their name + company]

**Note:** Common Room has limited data on this contact. No activity history, scores, or Spark profile available. I can run deeper web searches or look up their company for additional context.
```

Do not generate conversation starters, persona inferences, or engagement assessments from sparse data. These require real signals.

## Quality Standards

- Lookup must use the correct method for the input type — don't guess on email vs. handle
- Scores as raw/percentile only — never labels
- `Contact Initiated` activity (last 60 days) is the primary engagement signal — lead with it
- If Spark is unavailable, say so — don't fabricate a persona from title alone
- Flag any contact where the most recent activity is older than 30 days

## Reference Files

- **`references/contact-signals-guide.md`** — full field descriptions, Spark persona guide, and conversation starter principles

Referenced files: 4

View saved version →

---
name: prospect
description: "Build targeted account or contact lists using Common Room's Prospector. Triggers on 'find companies that match [criteria]', 'build a prospect list', 'find contacts at [type of company]', 'show me companies hiring [role]', or any list-building request."
---

# Prospecting

Build targeted account and contact lists using Common Room's Prospector. Supports iterative refinement through natural conversation, intent-based discovery, and both net-new prospecting and signal-based queries against existing accounts.

## Data Source

Use the Common Room MCP server and its object catalog. If its tools are unavailable or unauthenticated, tell the user that the Common Room connection is required and stop before producing a prospect list. Show only fields returned by the selected object type.

## Critical Distinction: Two Object Types

Common Room's Prospector operates against two fundamentally different object types. Always clarify which one is in play before running a query:

**`ProspectorOrganization`** — Companies **not yet in Common Room**
- Net-new companies that match specified criteria
- Available fields are firmographic only: name, domain, size, industry, capital raised, annual revenue, location
- Fewer filter options — no signal-based filters, no scores, no activity history
- Use when: building a brand-new target list, territory planning, top-of-funnel expansion

**`Organization`** (in Common Room) — Companies **already in your CR workspace**
- Full signal data available: product usage, community activity, CRM fields, scores, custom fields
- Much richer filter set — includes signal-based, score-based, segment-based, and firmographic filters
- Use when: finding warm accounts to prioritize, identifying expansion candidates, surfacing intent signals within existing pipeline

When a user's request could apply to both (e.g., "Show companies hiring AI engineers this month"), clarify:
> "Are you looking for net-new companies not yet in Common Room, or filtering accounts already in your workspace?"

The catalog should make this distinction explicit so the LLM can select the right Prospector endpoint.

## Step 0: Load User Context (Me)

Fetch the `Me` object to get the user's segments. When prospecting against `Organization` records (accounts already in CR), default to filtering within "My Segments" unless the user asks for a broader search.

## Step 1: Gather Targeting Criteria

If criteria are already provided, proceed. Otherwise ask:

> "What kind of accounts or contacts are you looking for? For example: company size, industry, job titles, signals like recent product activity or community engagement, geographic region, or specific intent signals like recent funding or job postings."

Use the Common Room object catalog to see available filters for each object type. The key distinction:
- **ProspectorOrganization** — firmographic and technographic filters only (industry, size, geography, funding, tech stack)
- **Organization** — all firmographic filters plus signal-based, score-based, segment-based, and CRM filters

**Lookalike search:** If the user asks to "find companies like [X]", first look up the reference company in Common Room (or via web search if not in CR). Extract its key attributes — industry, employee range, tech stack, funding stage, geography — and propose those as filter criteria. Present the derived criteria to the user for confirmation before running the search, since lookalike targeting works best when the user can refine which attributes matter most.

## Step 2: Support Iterative Refinement

Prospecting is conversational. Support multi-turn refinement naturally:

1. Run initial query with provided criteria
2. If results are large (50+), summarize and offer: "I found [N] results. Want to narrow by [suggested filter]?"
3. If results are too few (< 5), suggest: "Only [N] results with those filters — I can broaden by relaxing [specific criterion]."
4. Apply each refinement as a follow-up query, not a new search from scratch

Example flow:
- Rep: "Find cybersecurity companies in California." → 500 results
- Rep: "Only show ones over 300 employees using AWS." → 47 results
- Rep: "Focus on the ones with recent hiring activity." → 12 results ✓

## Step 3: Run the Query and Present Results

Execute the Prospector query with confirmed criteria. Sort by signal strength or fit score where available (not alphabetically).

**For `ProspectorOrganization` (net-new) results:**

| Company | Domain | Industry | Size | Capital Raised | Revenue | Location |
|---------|--------|----------|------|---------------|---------|----------|

**For `Organization` (in CR) results:**

| Company | Industry | Size | Top Signal | Signal Date | Score | CRM Stage |
|---------|----------|------|-----------|-------------|-------|-----------|

Flag any results where data is thin or the most recent signal is older than 90 days.

## Step 3.5: Enrich Net-New Results with Web Search

For `ProspectorOrganization` results (net-new companies not in CR), run a quick web search on the top 3–5 companies to add context beyond firmographics. CR has no behavioral signals for these companies, so web search fills the gap — look for recent funding, product launches, leadership changes, or news coverage. Include findings as brief annotations next to each company in the results.

## Step 4: Offer Next Steps

- "Want me to draft outreach for the top 3–5 prospects?"
- "Should I run a full account brief on any of these?"
- "Want to refine the criteria or add another filter?"
- "I can format this as a CSV if you'd like to export it."
- "For any net-new companies here, I can add them to Common Room for enrichment." *(future capability)*

## Quality Standards

- Always confirm which object type (ProspectorOrg vs Organization) before running the query
- Default to "My Segments" when querying Organization records, unless user specifies otherwise
- Support iterative refinement — treat each follow-up as a filter adjustment, not a fresh start
- Never mix result fields from ProspectorOrganization and Organization in the same list
- Fewer high-quality results beat a long unqualified list
- **Only show data the query returned** — leave blank or "—" for missing fields, don't invent values

## Reference Files

- **`references/prospect-guide.md`** — filter types, signal-based sorting, object type distinctions, and list-building strategies
- **`references/me-context.md`** — current-user and My Segments scoping

Referenced files: 5

View saved version →

---
name: weekly-prep-brief
description: "Generate a comprehensive weekly briefing for all external calls in the next 7 days. Triggers on 'weekly prep brief', 'prepare my week', 'what calls do I have this week', 'Monday prep', or any weekly planning request."
---

# Weekly Prep Brief

Generate a single comprehensive weekly briefing that covers every external customer or prospect call in the next 7 days, with per-meeting account and contact research from Common Room.

## Data Sources

Use the Common Room MCP server as the primary account and contact source. If its tools are unavailable or unauthenticated, tell the user that the Common Room connection is required and stop before producing signal-backed briefings. Use an available calendar connector to retrieve meetings; otherwise collect meeting details from the user.

## Briefing Process

### Step 1: Get the Week's External Meetings

**Option A — Calendar connected:**
Use an available calendar connector to fetch all meetings scheduled in the next 7 days (or a user-specified range). Filter to keep only external meetings — those with attendees from outside your organization. Discard internal-only meetings, one-on-ones with colleagues, and recurring internal syncs.

Identify for each external meeting:
- Company name
- Meeting date and time
- External attendee names and email addresses

**Option B — No calendar connected:**
Ask the user: "To build your weekly prep brief, I'll need your upcoming external calls. Please list them: company name, date/time, and attendee names."

Accept freeform input and parse it into a structured list before proceeding.

### Step 2: Confirm the Meeting List

Present the identified meetings to the user for confirmation before beginning research:

> "Here are the external calls I found for this week. Let me know if anything's missing or should be excluded:
> - [Company] — [Day], [Time] — [Attendees]
> - ..."

This prevents wasted research on cancelled or incorrect meetings.

### Step 3: Research Each Meeting

For each confirmed external meeting, run in parallel where possible:
1. **Account research** — full account snapshot using the account-research skill
2. **Contact research** — profile for each external attendee using the contact-research skill

Common Room data is the primary source. After CR research, run a quick **recency check** for each company — this is supplementary, not primary:
- Search `"[company name]" news` scoped to the last 7 days
- For executive attendees, search their name for recent public posts or interviews
- Only include findings that are genuinely noteworthy (funding, leadership changes, major press). Don't pad the brief with generic news.

Depth calibration:
- For high-priority accounts (large accounts, open opportunities, renewal risk), produce full depth research
- For lower-priority or short meetings, produce abbreviated snapshots (3–4 bullets each)

### Step 4: Synthesize the Weekly Brief

Compile all per-meeting research into a single structured document, sorted by meeting date/time.

Open with a brief week-level overview that flags:
- Any accounts with urgent signals (at-risk, trial expiring, expansion opportunity)
- Any meetings that need special preparation or executive involvement
- Total external call count and estimated time commitment

## Output Format

```
# Weekly Prep Brief — Week of [Date]

## Week Overview
[2–4 bullets: key themes, flagged priorities, call count]

---

## [Monday / Tuesday / etc.]

### [Company Name] — [Time]
**Attendees:** [Names and titles]
**Meeting type:** [Discovery / QBR / Renewal / Expansion / etc. — inferred if possible]

**Company Snapshot**
[4–5 bullets: account status, top signals, recent activity]

**Attendee Profiles**
- **[Name]** ([Title]): [2–3 bullets on their signals, persona, conversation angle]
- [Repeat per attendee]

**Top Signals This Week**
[2–3 most relevant signals for this specific call]

**This Week's News** [If notable news found]
[Only genuinely noteworthy findings — funding, leadership changes, major press]

**Recommended Objectives**
[1–2 sentences: what to accomplish in this meeting]

---

[Repeat per meeting, sorted by date/time]
```

## When a Meeting Has Sparse Data

If Common Room returns limited data for a particular meeting's account or attendees, use a compressed format for that meeting instead of the full template:

```
### [Company Name] — [Time] ⚠️ Limited Data
**Attendees:** [Names and titles if known]
**Data available:** [What Common Room actually returned]

**Web Search Results**
[Findings from web search — company news, attendee LinkedIn profiles]

**Note:** Common Room has limited data on this account. The rep may want to check directly in CR or gather context from colleagues before this call.
```

Do not generate a full meeting prep section (company snapshot, signal highlights, talking points, recommended objectives) from sparse data. A short honest section is more useful than a fabricated full one.

## Quality Standards

- Keep each meeting section scannable — reps read these in the morning, often on mobile
- Always sort by date/time ascending
- Flag urgent situations prominently (risk, trial expiration, open opps) — don't bury them
- If a meeting has very thin Common Room data, use the sparse-data format above — never fill the full template with guesses
- Total brief should be readable in 10–15 minutes for a week with 4–6 meetings
- **Every fact must come from a tool call** — no invented deal context, activity, or signals

## Reference Files

- **`references/briefing-guide.md`** — guidelines for structuring briefings, prioritization logic, and how to handle edge cases (cancelled meetings, new accounts with no data, etc.)
- **`references/me-context.md`** — current-user and My Segments scoping

Referenced files: 5