Query Clari forecast and deal-risk data from Claude before forecast calls
Before a forecast call, this assistant pulls the latest forecast, recent deal changes and risk scores so a revenue leader can ask plain questions about them. It saves the weekly export-and-paste routine.
Paste this into ChatGPT, Claude or your agent tool. See the prompt
- Deal Review / Forecasting
- advanced
- RevOps
- Sales leader
- Category
- Deal Review / Forecasting
- Difficulty
- Advanced
- Est. cost
- Not stated by source
Works with
Problem
RevOps opens Clari every week, exports a forecast slice and pastes it somewhere to answer a question they already know the shape of. Cross-checking risky deals against the CRM adds more steps.
Setup
A small read-only MCP server gives Claude three tools against the Clari API: get the forecast, list pipeline changes such as stage moves, amount edits and close-date pushes, and get AI deal-risk scores. Before a forecast call, Claude pulls the forecast, surfaces flagged deals and cross-references the week's pipeline events, optionally alongside a Salesforce MCP server in the same session.
Prompt
Starter prompt by The GTM Agent Weekly, written from the setup described on this page. It is not the source's own prompt; see the source link below for their write-up.
# ROLE
You help a sales leader prepare for the weekly forecast call. You are read-only.
# TOOLS
- {{FORECAST_TOOL}}: get the current forecast by team or rep (e.g. Clari).
- {{PIPELINE_CHANGES_TOOL}}: stage moves, amount edits and close-date pushes this week.
- {{DEAL_RISK_TOOL}}: deal-risk scores and reasons.
- Optional {{CRM_TOOL}}: opportunity details and activity (e.g. Salesforce).
# TASK
1. Pull the forecast for {{TEAM}} for {{PERIOD}}: commit, best case, pipeline, versus last week.
2. List the week's pipeline changes. Group them: pushed close dates, amount changes, stage
moves forward, stage moves back.
3. Pull deals flagged as at risk. Cross-check each against this week's changes and the last
activity date.
4. Pick the deals that most need discussion: large commit deals that slipped, got riskier or
went quiet.
# OUTPUT FORMAT
FORECAST: commit, best case, change vs. last week.
WHAT MOVED: grouped bullets with deal, owner and the change.
DISCUSS THESE: up to {{MAX_DEALS}} deals, each with why and one question to ask the owner.
# GUARDRAILS
- Read-only. Never change a forecast, stage or amount.
- Every number comes from a tool call. If a tool fails, say which and leave that section empty.
- Do not predict the final number yourself. Replace every {{PLACEHOLDER}} before running. Works with any capable LLM and runner.
Tools
Steps
- Install the server in a Python virtual environment.
- Generate a Clari API token from a dedicated service account with a minimal role.
- Copy the forecast ID from the Forecast tab URL into configuration.
- Register the server in Claude Desktop and ask a pre-call question such as which deals slipped this week.
- Run it next to a Salesforce MCP server to compare Clari risk flags with CRM stages.
Results
Workflow bundle with documentation; no outcome data.
What goes wrong
Gotchas
- The forecast API is asynchronous, so each forecast call takes seconds of polling; fine for prep, not for a live call.
- If Clari's CRM sync lags, Claude describes a stale picture.
- Deal values and contact data pass through the LLM, so clear your AI data policy first.
Get 5 like this every week, free.
Real agent setups with sources and failure modes, in one short weekly email.
Free. One email a week. Unsubscribe anytime.
Source and credit
ooligo · vendor
Source dated 2026-10-11 · last reviewed 2026-10-11. Summarized in our own words; see the source for details.