hermes

Write a data request brief

Turns a vague stakeholder ask into a clear data request with the decision, exact metric definitions, filters, time range, format and deadline, plus open questions. Use when a data ask arrives vague.

context

You sit between business teams and the data team. "Can you pull the numbers on churn for last quarter?" can mean twenty different queries, and the analyst usually guesses one, delivers it a week later, and starts again. A good brief fixes the ambiguity in five minutes: it names the decision, pins down every definition, and lists exactly what will be delivered and when.

task

Turn this ask into a data request brief.

ask

context

  1. Infer the decision or use behind the ask (a board slide, a pricing decision, a campaign review). If it is not stated, propose the most likely one and mark it "to confirm".
  2. Rewrite the ask as precise questions, each answerable with one table or chart.
  3. For every metric, write a definition: formula, unit, what is included and excluded (test accounts, refunds, internal users, free plans), grain, and time zone. Where a common term is ambiguous ("active users", "revenue", "churn", "conversion"), give the two or three plausible definitions and recommend one.
  4. Fix the population and filters, time range (with exact dates, and whether the current incomplete period is included), comparison (previous period, last year, target), and breakdowns.
  5. Specify the deliverable: format (number in a message, table, chart, dashboard, spreadsheet), level of detail, and who receives it.
  6. Set the deadline and priority, and the smallest useful version that could be delivered sooner.
  7. List the questions that must be confirmed before work starts, at most five, ordered by how much they change the result.
  8. Write a short reply message to the requester that confirms the brief and asks those questions.
constraints
  • Mark every inference "to confirm"; do not present guesses as agreed.
  • Do not produce any numbers or results.
  • Keep the brief to what fits on one screen. Use the requester's vocabulary and avoid jargon in the reply message.
  • If the ask contains requests for personal data about individuals (for example a list of named customers), note whether aggregate data would serve the purpose and flag data-access approval if needed.
output format

Brief

A table: field | value. Fields: Requester, Decision or use, Questions, Metric definitions, Population and filters, Time range, Comparison, Breakdowns, Deliverable, Deadline and priority, Smallest useful version, Known caveats.

Questions to confirm

A numbered list, at most five.

Reply message

A short message ready to send to the requester.

1 required value still a placeholder; the assistant will ask for it.

details

kind
Prompt: a task you run by name to get one finished thing back
domain
Data analysis
category
Data exploration
level
Beginner
made for
Data analyst, Product manager, People manager, Marketer
risk
read-only
version
v1.0.1 · incubating
reviewed
2026-10-02
works in
Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Antigravity, OpenCode, Windsurf, Zed, Continue, AGENTS.md, ChatGPT, claude.ai

Edit on GitHubReport a problem

use in

Hodios CLI
npx @hermes-hq/hodios install write-data-request-brief --target claude-code
Agent Skills
npx skills add hermes-hq/hodios-dist --skill write-data-request-brief -a claude-code
Add the Hodios marketplace (once)
claude plugin marketplace add hermes-hq/hodios-dist
Install the data-analysis plugin
claude plugin install hodios-data-analysis@hodios

The plugin brings every entry in this domain at once.

PromptData exploration

Write an analysis plan

Writes an analysis plan before touching data, covering the decision, questions, metrics, data, method, comparisons, pitfalls and the result that would change the decision. Use when scoping a request.

write-analysis-plan
PromptReporting

Define a metric

Writes a precise metric definition (formula, grain, filters, edge cases, owner, known caveats) so every team computes the number the same way. Use when a metric is disputed or about to be launched.

define-metric
PromptData exploration

Answer a question with SQL

Turns a business question and a schema into an analytical SQL query, states the assumptions behind it and explains how to read the result. Use when you know the question but not the query.

answer-question-with-sql
WorkflowReporting

Analysis project track

Takes a stakeholder request from question to analysis plan, data checks, analysis and a decision-ready report, pausing for review between steps. Use when an analyst takes on a request.

analysis-project-track
PersonaData exploration

Data analyst

Acts as a data analyst who starts from the decision, sanity-checks data before trusting it and states uncertainty plainly. Use as a standing analyst persona or subagent for data questions.

data-analyst
PromptData exploration

Reconcile two datasets

Reconciles two datasets that should agree, such as bank versus ledger or CRM versus billing, by matching records, listing mismatches and explaining likely causes. Use for month-end checks.

reconcile-datasets