name: client-intelligence-brief description: > Turn a pile of recent client correspondence into a structured intelligence brief in under 3 minutes. Paste 3-5 emails — from the client, to the client, or session recaps — and get back: a full state-of-play analysis, a call prep with specific openers and watchpoints, and a prioritized action plan. Use when you want to prep for a client call, catch up after being heads-down, or hand off client context to a team member without a 20-minute download. Triggers: "client brief", "prep for client call", "what's going on with this client", "catch me up", "client intelligence brief", or any request to synthesize client communications into actionable status. metadata: author: Kathryn Brown, Advisory OS version: "3.1.0" updated: "2026-04-16"

Client Intelligence Brief

Turn recent client communications into the brief you wish someone on your team produced for you before every call. Paste emails, walk in prepared.

Core Principle

Read before writing. Absorb all the emails first. Don't start producing output until you've read every message the user provides. The quality of the brief depends on seeing the full picture before drawing conclusions.

What This Skill Does

You paste 3-5 recent emails related to a single client — emails from them, emails to them, session recaps, or any mix. The skill reads them and produces a brief that does three jobs:

Job 1: State of Play — What's happening across every active workstream. Not a summary. A structured read of what's moving, what's stuck, what's at risk, what the client cares about most right now, and how the client is feeling about their own business and about working with you. The kind of analysis a senior advisor would produce after reading the same emails — the patterns you'd catch if you had 30 minutes to think, delivered in 2.

Job 2: Call Prep — How to open the conversation, what to press on, what to leave alone, what to listen for, and what to lock down before hanging up. Not theory. Actual language you can use. The goal is to walk in with a plan, not wing it based on whatever you remember.

Job 3: Action Plan — What to do this week, ranked by priority, with specific rationale tied back to the analysis. Not "follow up with client" — what to say, why now, and what signal from the brief is driving it.

The Brief: Section by Section

1. Orientation

Two to three sentences. Who is this client, what phase is the engagement in, what's the overall cadence. Just enough context so someone picking this up cold knows where they are. If you're using this yourself, this is your mental reset — clear out whatever you were doing before and lock in on this client.

2. Rolling Items Alert

Surface this FIRST, before the full open items list. This is the single highest-value feature of the brief — it catches the things that keep slipping without anyone noticing.

Count how many items have appeared as action items in one email and remain unresolved in a subsequent email. Display as a bold count:

X items rolling across multiple emails.

Then list each one:

• [Item] — first assigned [date/email], still open. Owner: [who]. Now [X weeks/emails] without resolution.

If nothing is rolling, say: No rolling items detected.

3. Open Items by Workstream

Every unresolved question, pending deliverable, or waiting-on from either side, grouped by the distinct workstreams visible in the emails. Each workstream gets a short label and its own checklist.

Why workstream grouping matters: A flat list of 12 open items feels overwhelming and hides the real picture. Grouping reveals that three workstreams are moving fine and one is completely stalled. That changes how you spend your time on the call.

Rules:

• Label each workstream with a clear, short name
• Mark who owns each item (client, you, joint, or third party)
• Include deadlines or timing if mentioned in the emails
• Items that also appear in the Rolling Items Alert get tagged [Rolling]

4. Velocity

Are action items from earlier emails showing up as completed in later ones? This is the single most important signal for whether the engagement is working.

Look at what was assigned vs. what got done across the email sequence. Call the pattern explicitly:

• Strong execution — most items completing between sessions
• Selective execution — some types of work completing, others consistently deferred (name the pattern — what gets done vs. what doesn't)
• Rolling — items carrying forward across multiple emails without completion

Format: Two-column table. Numbers, Pattern, Completing, Deferring, Read — one row each. No paragraphs, no wrapping prose.

The Numbers row is critical. Count actual items: "4 of 7 completed. 2 rolling (1 for the second time). 1 new." Numbers make the time-saving tangible — the skill just counted what the user would have spent 10 minutes reconstructing from memory.

Important: Velocity requires at least two emails in sequence to assess. If only one email is provided, say so — don't guess.

5. Client Priority Read

What does the CLIENT care about most right now? Not what you think they should care about — what are THEY focused on based on their emails?

Read the emails for what they bring up first, what they spend the most words on, what they ask questions about, what they return to across multiple emails. Distill it to one phrase.

Format: Two-column table. "Their #1 right now" in one row, "Evidence" in the next.

Why this matters: If you can name their priority in the first 30 seconds of the call, they feel heard. If you lead with YOUR priority instead, you start on the back foot. This tells the user where to meet the client.

6. Sentiment: Their Business

How does the client feel about their own work, their clients, their team, their market position? This is not about you or the engagement — it's about what's going on in their world.

Format: Two-column table. Temperature (one word) in one row, Evidence (1-2 lines of specifics) in the next. No paragraphs. Cite specific language, situations, or tone shifts. "Client mentioned two previously siloed departments are now meeting weekly" is evidence. "Client seems positive" is not.

Why this matters for call prep: If the client just landed a big win, you open differently than if they're navigating a political mess. This section tells you what emotional state you're walking into.

7. Sentiment: This Engagement

How does the client feel about working with you? Are they engaged, overwhelmed, pulling away, gaining confidence, going through the motions?

Different signal, different response required. Someone who's energized needs momentum. Someone who's overwhelmed needs you to reduce scope. Someone who's pulling away needs direct conversation. This section tells you which mode to operate in.

Format: Two-column table. Temperature in one row, Evidence in the next. No paragraphs. Look for: session attendance patterns, tone in email responses, whether they're volunteering information or only answering questions, whether action items are completing (engagement signal) or piling up (overwhelm or disengagement signal).

8. Risks & Watch Items

Things that aren't problems yet but could become problems. This is where the brief earns its keep — surfacing what you'd miss in a quick email scan.

Patterns to look for:

• Rolling items: Anything flagged in Section 2. If items are rolling, they appear here too with the risk framing — what happens if they keep slipping.
• Capacity signals: Total count of open items across all workstreams. If a client has 10+ open items, that's a capacity risk regardless of how engaged they seem.
• Dependencies: Items where progress on one workstream requires completion of something in another. If the dependency is stuck, the downstream work is at risk.
• External pressures: Client-side events (acquisitions, leadership changes, seasonal crunches, team departures) that could redirect their attention away from engagement work.
• Silence on a topic: Something discussed actively in earlier emails that disappears from later ones. Silence sometimes means it's handled. Sometimes it means avoidance.

If nothing is developing, say "No emerging risks identified" — don't manufacture them.

9. Call Prep

This is the section that makes the brief worth more than a summary. Five components:

Open with: The specific language for the first 30 seconds of the call. Not "ask about their week." A concrete opener that demonstrates you know exactly where things stand and what matters right now. Structure: acknowledge something specific from the last interaction, then bridge to what you want to focus on today.

Example format: "Last time we [specific thing]. I want to check on [specific open item] before we get into [priority topic]."

Press on: One to two items that need direct attention this call. These are the items where delay creates risk or where the client has been deferring without naming why. For each one, include what to say and how to frame it — not confrontational, but clear enough that it can't be deferred again.

Leave alone: Items that are tempting to raise but shouldn't be touched this call. Maybe the client is already maxed on action items. Maybe something is politically sensitive and the timing is wrong. Maybe it's progressing fine and bringing it up signals distrust. Name the item and say why you're leaving it.

Listen for: Signals during the call that would update the brief. Things the client might say that would confirm a risk, reveal a blocker they haven't named, or change priority. These are the things you'd catch if you were fully present instead of half-scanning your notes trying to remember what's open.

Confirm before ending: The one or two things to lock down before hanging up. Typically: get a commitment on the rolling item you pressed on, and confirm the next session date/focus. This isn't a scripted close — it's a reminder of what can't leave the call unresolved. One line.

10. Next Actions

What to do this week, ranked by priority. Each action gets two lines:

• Bold action name
• Why now: [the signal from the brief]
• Do this: [what specifically to do or say]

Three to five actions maximum. If the brief surfaces more than five, prioritize ruthlessly — everything can't be urgent. Rank by: what prevents the most damage first, what creates the most momentum second, what maintains the relationship third.

Quality Check (Internal — never shown to the user)

This section is an internal gate. Run it silently before presenting. Use it to correct the brief. Do not include the Quality Check in the output. The user should never see a Pass/Fail table or a "weakest section rewrite" note — they should see a finished brief that has already been through the gate.

Before presenting, verify the brief against four checks internally:

Enforcement rules:

1. Failed checks: If any section fails a check, fix it before presenting. The output contains only the corrected section — no flag, no note about what was fixed.

1. Weakest section: Identify internally which section of the brief is weakest and rewrite it before presenting. Verify the rewrite internally: quote the original line(s) and the replacement line(s) to yourself, then confirm the replacement text is actually in the final brief. If it isn't, the rewrite didn't happen — redo it.

1. Missing data: If the emails don't contain enough information to assess a section, note it in the brief itself: "Not enough data to assess [section]." A flagged gap is honest and user-facing. A fabricated assessment is a failure.

What the user sees: The finished brief. No Quality Check table. No weakest-section note. No commentary about what the skill evaluated or changed.

Rules

• No paragraphs in the output. Every section uses tables, label/value pairs, bullet points, or single-sentence lines. The brief is a pre-call scanning tool, not a document to read. If any section requires more than a 5-second glance to absorb, it's too dense. Break it up.
• One client per brief. Don't mix emails from multiple clients in one run.
• Recency matters. Use emails from the last 30 days for the most accurate read. Older emails give historical context but the brief should reflect NOW.
• Include both sides when possible. Emails from the client AND emails/recaps you sent give the skill the full picture — what you've already addressed vs. what's still open. A brief built from only one direction misses half the context.
• Don't assume. If the emails don't contain enough information to assess sentiment, velocity, or risk, say so. "Not enough data to determine X" is better than a guess. One email is enough for open items and orientation but not for velocity or pattern detection.
• No client names in output unless the user provides them. If the user pastes emails with names, use them. If they've anonymized, keep it anonymous.
• Keep the brief scannable. Each section should be absorbable in a single glance — 5 seconds, not 15. The entire brief should take under 2 minutes to scan. Use bold labels, short lines, and white space. Cut adjectives, compress sentences, remove anything that doesn't change what you'd do on the call.
• [Rolling] tags are mandatory. Any item that appears as an action item in one email and remains unresolved in a subsequent email gets tagged [Rolling]. These also surface in the dedicated Rolling Items Alert (Section 2).
• Call prep language should be usable as-is. The "open with" section should be something the user can say verbatim on the call. Write it in first person, in natural conversational language, not in consultant-speak.
• Numbers over labels. Wherever possible, use concrete counts (3 of 5 completed, 2 rolling) instead of qualitative labels alone. Numbers make the analysis feel rigorous and the time savings tangible.
• Output as a markdown file. Always produce the brief as a .md file, not inline chat. Name the file: [client-name]-intelligence-brief-[YYYY-MM-DD].md (lowercase, hyphens, no spaces). The brief is a reference document the user pulls up during the call — it needs to render cleanly in the artifact panel with proper tables and formatting.

Output Format

# Client Intelligence Brief

| | |
|---|---|
| **Client** | [Name or identifier] |
| **Date** | [Today's date] |
| **Based on** | [X emails, date range] |
| **Next call** | [Date if mentioned, otherwise "Not specified"] |

---

## Rolling Items Alert

**[X] items rolling across multiple emails.**

- **[Item]** — first assigned [date/email], still open. Owner: [who]. [X weeks/emails] without resolution.
- **[Item]** — first assigned [date/email], still open. Owner: [who]. [X weeks/emails] without resolution.

---

## Open Items

### [Workstream 1]
- [ ] [Item — owner — deadline] [Rolling]
- [ ] [Item — owner]

### [Workstream 2]
- [ ] [Item — owner]

### [Workstream 3]
- [ ] [Item — owner]

---

## Velocity

| | |
|---|---|
| **Numbers** | [X of Y completed. X rolling (details). X new this cycle.] |
| **Pattern** | [Strong execution / Selective execution / Rolling] |
| **Completing** | [What types of work are getting done] |
| **Deferring** | [What types keep slipping — or "Nothing notable"] |
| **Read** | [One sentence: what the pattern means] |

---

## Client Priority Read

| | |
|---|---|
| **Their #1 right now** | [One line — the thing they're spending the most energy on] |
| **Evidence** | [Specific language, questions they keep raising, topics they lead with] |

---

## Sentiment: Their Business

| | |
|---|---|
| **Temperature** | [One word] |
| **Evidence** | [1-2 lines, specific] |

## Sentiment: This Engagement

| | |
|---|---|
| **Temperature** | [One word] |
| **Evidence** | [1-2 lines, specific] |

---

## Risks & Watch Items

- **[Risk]** — [What's happening, why it matters, what to monitor]
- **[Risk]** — [What's happening, why it matters, what to monitor]

---

## Call Prep

**Open with:** "[Specific language for the first 30 seconds.]"

**Press on:**
- [Item] — [How to frame it, what to say]

**Leave alone:**
- [Item] — [Why not this call]

**Listen for:**
- [Signal] — [What it would mean]

**Confirm before ending:** [What to lock down before hanging up — typically the rolling item commitment and next session focus]

---

## Next Actions

**1. [Action]**
Why now: [Signal from the brief]
Do this: [What specifically to do or say]

**2. [Action]**
Why now: [Signal from the brief]
Do this: [What specifically to do or say]

**3. [Action]**
Why now: [Signal from the brief]
Do this: [What specifically to do or say]

What Makes This Different

Most "summarize my emails" tools give you a summary. This gives you a call prep.

The difference: a summary tells you what happened. This brief tells you what to do about it — what to say when the call starts, what to push on, what to leave alone, what to listen for, and what to lock down before hanging up. It catches the rolling items nobody notices, the sentiment shifts that change how you show up, and the risks that aren't problems yet but will be next month if you don't act now.

You walk in knowing what the client cares about most right now, not guessing. Two minutes replaces the 15-minute scramble.

The brief is built for people who manage 10-30 client relationships and can't hold it all in their head. This is the visibility you've been doing mentally. Now it's structured, consistent, and catches what you'd miss.