Campfire Analyst
For Users: How to Use This Skill
What it does
Campfire Analyst reads your Zerodha portfolio and produces a complete picture of your investments — what you own, how it's performing, what's broken, and what your money reveals about you as a person.
It generates three files:
- campfire-dashboard.html — four-tab visual app: Summary (headline + quick wins), Fix List (task center), Analysis (Vitals + portrait + sectors), and X-Ray (look-through inside your mutual funds)
- investment.md — a detailed wealth document you can save or share with an advisor
- taste.md — a lifestyle identity profile built through a prediction game
How to start
Just type /campfire-analyst in the chat. That's it. No setup, no options to configure.
You'll be asked to log into your Zerodha account via a one-click link. Once authorized, the analysis runs automatically.
What happens next
- You'll see the Campfire Dashboard — a four-tab HTML file. The Summary tab leads with a single headline finding and the three most actionable quick wins. The Fix List tab has the full portfolio task center. The Analysis tab shows Campfire Vitals, style portrait, thematic map, concentration risk (combined across direct equity and MFs), and benchmark comparison. The X-Ray tab shows the full breakdown of what's inside each mutual fund — the evidence behind the concentration numbers, loaded automatically on page open.
- investment.md follows immediately — the full wealth document with deep allocation tables.
- Then the prediction game begins — Claude tries to guess your phone, car, watch, hotel preference and more from your portfolio alone, one prediction at a time. At the end it generates taste.md.
If you only want part of it
You can ask for specific pieces directly:
- "Just give me my Campfire Vitals report"
- "Generate my investment.md"
- "Let's do the prediction game"
- "Show me my fix list"
- "Run Portfolio X-Ray"
If you don't use Zerodha
You can still run the analysis manually. Share your holdings as a screenshot, CSV, or just paste your top 20-30 positions with quantities and average buy prices. Claude will work with whatever you provide.
What it doesn't do
It doesn't execute any trades, place orders, or modify your account in any way. It only reads your holdings data.
For Skill Authors: Framework Reference
A seven-part portfolio intelligence framework that transforms raw holdings data into a complete financial + consumer identity profile.
When triggered, this skill runs seven sequential steps on the user's portfolio:
| # | Analysis | What It Produces | Reference File |
|---|---|---|---|
| 1 | Deep Portfolio Analysis | Sector breakdown, winners/losers, asset class mix, MF breakdown, thematic map, expense audit, liquidity, yield, beta | (computed internally — feeds dashboard) |
| 2 | Campfire Vitals Report | 8 custom financial health markers scored 0-100 | references/vitals-metrics.md (computed internally — feeds dashboard) |
| 3 | Investment Style Portrait | Written paragraph characterising the investor's style and blind spots | (computed internally — feeds dashboard) |
| 4 | Portfolio X-Ray Classification | Fund tier classification + JavaScript engine for browser-side look-through | references/xray-template.md (embedded in dashboard HTML) |
| 5 | Campfire Dashboard | Four-tab HTML file — Summary · Fix List · Analysis · X-Ray | references/dashboard-template.md, references/task-center-template.md, references/xray-template.md |
| 6 | investment.md Generation | Complete wealth document with deeper consolidated allocation tables | (file output) |
| 7 | taste.md Consumer Identity | Archetype extraction → lifestyle prediction game | references/archetypes.md, references/prediction-chains.md |
CRITICAL: Every analysis must be derived fresh from THIS user's data. Do not carry over any assumptions, archetypes, predictions, or personal details from any prior conversation. Every user is unique. Every portfolio tells a different story.
Step 0: Pull Data
Primary: Kite MCP
kite:get_holdings → All stock/ETF positions
kite:get_mf_holdings → All mutual fund positions
kite:get_margins → Cash balance
kite:get_positions → Intraday/F&O positions (optional)
If kite:get_mf_holdings fails or returns empty, note this and proceed with direct equity only. Flag the gap.
Fallback: Manual Upload
If Kite MCP is not connected, ask for:
- Holdings screenshot or CSV export
- Mutual fund statement (CAS PDF or broker export)
- Or at minimum: top 20-30 holdings with quantities and average buy prices
Data Preparation
Once data is pulled, compute for every holding:
- Current value = quantity × last_price
- Invested value = quantity × average_price
- P&L = current - invested
- Return % = (last_price - average_price) / average_price × 100
Compute portfolio totals:
- Total invested, total current value, total P&L, overall return %
- Cash balance from margins
- Total wealth = direct equity + MF + cash
Part 1: Deep Portfolio Analysis
Do not present this section inline. Compute all values silently. Everything here feeds directly into the Campfire Dashboard (Part 5), which is the first and only time this data is shown to the user.
1.1 Portfolio Snapshot Table
| Metric | Direct Equity | Mutual Funds | Combined |
|---|---|---|---|
| Invested | |||
| Current Value | |||
| Unrealised P&L | |||
| Return % | |||
| Holdings Count |
1.2 Asset Allocation
Classify EVERY holding into asset types:
- Indian Equities (direct + MF equity)
- Gold & Silver (ETFs + SGBs + MF commodity funds)
- International (US/global ETFs + MF international funds)
- Index Trackers (Nifty/Sensex, direct + MF)
- Multi-Asset Funds
- ELSS Tax Saver Funds
- Arbitrage / Liquid
- Debt (if any)
- Cash
1.3 Sector Allocation (Direct Equity)
Classify every direct stock into: Banking & Finance, Energy & Utilities, Telecom, IT & Tech, Pharma & Healthcare, Consumer & FMCG, Auto & Industrial, Metals & Mining, Chemicals, Infra & Construction, Travel & Hospitality, Retail & Lifestyle, Conglomerate, Gold/Silver, Index & Liquid ETFs
Show value and % for each sector. Identify top 3 sector bets.
1.4 MF Category Allocation
Classify every mutual fund into: Flexi Cap, Nifty 50/Sensex Index, ELSS Tax Saver, Multi-Asset/Dynamic, Silver/Commodity, Arbitrage/Liquid, US/International, Small & Mid Cap, Thematic/Sectoral
Show value and % for each category. Identify AMC concentration (% of MF portfolio per AMC).
1.5 Key Tables
- Top 15 Positions by current value (stock name, value, return %, P&L)
- Multibaggers (100%+ returns) sorted by return %
- Biggest Losers sorted by absolute P&L loss
1.6 Consolidated Views
Build cross-cutting views that merge direct equity and MF data:
- Precious Metals (all vehicles) — every gold/silver instrument across direct + MF
- International Exposure (all vehicles) — every US/global instrument
- Nifty 50 Trackers (all vehicles) — every Nifty/Sensex instrument
These consolidated views reveal the TRUE allocation that individual portfolio views hide.
1.7 Concentration Risk (Direct Equity)
Compute concentration in direct equity holdings:
Top position concentration:
top1_pct = largest_holding_value / total_de_value × 100
top3_pct = sum of top 3 holdings / total_de_value × 100
top5_pct = sum of top 5 holdings / total_de_value × 100
top10_pct = sum of top 10 holdings / total_de_value × 100
Thresholds:
- Top 1 position > 15% → flag as HIGH concentration
- Top 3 positions > 35% → flag as HIGH
- Top 5 positions > 50% → flag as MEDIUM
- Top 10 positions > 70% → NORMAL for a focused portfolio
Stress scenario (compute silently, show as context): "If your largest position halve