Guide · Agent skill
How to use the Gather Business Context agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The Gather Business Context skill is an agent skill that makes an AI agent collect background before it starts an analysis. Agent skills are folders of instructions that an agent loads when a task matches them. This guide explains what the skill does, how to install it, how to trigger it with plain requests, and what to watch for around access and network use.
Gather Business ContextXiaomiMiMo/MiMo-Code · MIT Download the skill Quick start
- 1Download the ZIP. Download the skill ZIP from this page and unzip it.
- 2Place or upload the skill. Put the folder in ~/.claude/skills/ for personal use or .claude/skills/ for one project, or upload it under Settings → Capabilities → Skills in the Claude apps.
- 3Enable your sources. Turn on the warehouse catalogs, dashboards or docs the agent should search.
- 4Ask a context-dependent question. Describe the metric, change or source you need background on, with a time period if you have one.
- 5Review the context note. Check the summary, citations, labelled conflicts and listed gaps before starting the analysis.
- 6Continue to the focused skill. If you also need a diagnosis or deliverable, let the agent hand off to the matching focused skill.
What does the Gather Business Context skill make the agent do?
When an analytical question depends on missing background, such as what a metric means, what changed recently, or which sources should be checked, the skill tells the agent to search the connected or provided sources first. The agent then returns a short context note with citations, conflicts and gaps.
The skill is deliberately limited to framing. It does not run the analysis itself. If the same request also asks for a diagnosis, recommendation, dashboard or report, the skill says to return only the context needed for the next step and continue with a focused skill such as metric-diagnostics, product-business-analysis, build-dashboard or build-report.
- Identifies the topic, the time period and the decision the analysis supports
- Searches with concrete names, owners, dashboards and source names rather than broad guesses
- Prefers authoritative artifacts (closest to what was defined, decided or measured) over informal chatter
- Labels assumptions, conflicts and missing source-of-truth documents
Install the skill from the ZIP or the Claude apps
Download the ZIP for this skill from this page and unzip it. For personal use, place the unzipped folder in ~/.claude/skills/. To limit it to one project, place it in that project's .claude/skills/ folder.
If you use the Claude apps, you can upload the skill instead. Go to Settings, then Capabilities, then Skills, and upload the skill there.
The package contains agents/openai.yaml and SKILL.md. The skill is used in the MiMo-Code agent environment, so the agent must have the skill installed to use it.
- Personal: ~/.claude/skills/
- Single project: .claude/skills/ inside the project
- Claude apps: Settings → Capabilities → Skills
Trigger the skill with natural requests
You do not need a special command. Ask a question that depends on background the agent does not have, and the skill's description tells the agent when to use it. Phrases about what a metric means, what changed recently, or which source is current are the clearest triggers.
If your request also asks for a deliverable, say so in the same prompt. The skill is written to gather context first and then continue to the focused skill, so the framing step and the analysis step stay separate.
- "What does the activation metric mean in our reporting, and which dashboard is the current source of truth?"
- "What changed in checkout conversion recently? I need context before we investigate."
- "Which owner notes or decision docs should I check before analyzing the pricing test?"
- "Gather the context for the retention question, then diagnose the drop."
Know what the skill needs and what it does on your machine
The skill needs access to whatever data or document sources you have enabled, such as warehouse catalogs, dashboards or docs. The files do not name specific products, so the skill works with the sources your setup exposes. Without connected sources or user-provided context, the agent may guess at definitions, so supply the definitions yourself when you can.
The skill does not run scripts. It does use the network, because it searches your connected sources. Connected sources can be broad, and the skill can surface sensitive internal data if access is not scoped carefully. Enable only the sources you intend the agent to read.
- Runs scripts: no
- Uses network: yes (to search connected or provided sources)
- Be deliberate about which sources you enable, since broad access can expose sensitive internal data
Tips for getting useful context notes
Give the skill concrete anchors. A metric name plus a dashboard name, or a feature plus a launch window, narrows retrieval better than a general topic. If you know the time period, state it; otherwise the skill uses the narrowest reasonable window and labels it as an assumption.
Skip the skill when your prompt already contains the context. The skill itself says to skip it in that case.
- Name the metric, dashboard or owner you already know about
- State the time period and the decision the analysis supports
- Ask for a short orientation if you only need quick framing; the skill shortens its structure but keeps citations and conflicts
What people use it for
- Checking which dashboard or decision document is the current source of truth before a metric investigation
- Finding the agreed definition of a metric before comparing it across teams
- Collecting recent changes to a feature or rollout before analyzing its impact
- Separating the framing step from a request that also asks for a report or dashboard
Frequently asked questions
How do I install the gather-business-context skill?
Download the ZIP from this page and unzip it. Place the folder in ~/.claude/skills/ for personal use or in a project's .claude/skills/ folder for one project. In the Claude apps, you can instead upload it under Settings → Capabilities → Skills.
What do I need to use this skill?
You need access to the data or document sources you have enabled, such as warehouse catalogs, dashboards or docs, plus the MiMo-Code agent environment with the skill installed. The files do not name specific products.
Is the Gather Business Context skill free?
The source material for this skill is released under the MIT licence. This page does not state pricing for anything beyond that, so check the repository for licence terms.
Does the skill run scripts on my computer?
No. The skill does not run scripts, but it does use the network to search connected or provided sources.
Why did the agent skip the context step?
The skill says to skip gathering when the prompt already provides the needed context or the task is fully self-contained. Add the missing background to the prompt if you want the search to run.
What should I use instead if I need a full report?
This skill excludes broad background research and final reports. Use the focused skills it names, such as build-report or metric-diagnostics, for those deliverables.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you are an analyst or product team that needs the right definitions, recent changes and current source of truth before an investigation, and you want the framing step kept separate from the analysis. Skip it if your prompt already has the context, if you need a broad research summary or final report, or if your setup has no connected sources, because the agent may then guess at definitions.
Download the skill