Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/microsoft/data-formulator/data_loadingnpx skills add microsoft/data-formulator --skill data_loadinggit clone --depth 1 https://github.com/microsoft/data-formulatorWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00037 | $0.01231 |
| Opus 5 | $0.00018 | $0.00616 |
| Sonnet 5 | $0.00007 | $0.00246 |
| Haiku 4.5 | $0.00004 | $0.00123 |
Grade A, and why
data-loading scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
100% identical to data-loading — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: Data discovery
The workspace tables listed in your context are the data already loaded into the system, and the only data that can be read directly. Everything these tools return is not loaded yet — it lives in a connected source and only becomes usable after the user selects a loading option and the server materializes it.
Use these tools to determine whether connected sources contain data needed for the user's goal. They are read-only: discovering, describing, or probing a source does not add anything to the workspace analysis inputs.
Adding a connector
When the user wants to connect a new source, do not merely ask them to navigate to settings and do not attempt to connect on their behalf.
- Call
list_connectorsfirst because available built-ins and plugins vary by deployment. For a broad request such as "help me connect", summarize the concrete available types and ask which one they use. - Once the source type is known, call
describe_connectorwhen field or auth details are useful. - When the requested source type is known and available, you MUST call
propose_connectionin this same turn. Do not stop with text such as "I'll open the form", "you'll need to provide", or a list of required fields. Only the action opens the form. Include one or two helpful sentences alongside the action call explaining what the user should review or supply; this text appears above the chat while the form opens on the canvas. Passprefilledvalues the user already supplied, including values parsed from a connection string or config snippet. Never invent missing values. - The form is only a proposal. The user reviews it and clicks Connect; the action must never connect automatically.
Prefilled values may include credentials the user deliberately supplied. Do not repeat those values in prose or subsequent tool output. They are transient form seeds and are removed from persisted UI state.
Discovery sequence
- Use
find_datawhen the user names a business concept or table. Uselist_datawhen you need to browse available sources or hierarchy. - Use
describe_databefore relying on columns, types, row counts, or filter values. Pass the exactsource_idandtable_keyreturned by discovery. - Use
probe_dataonly when metadata is insufficient to choose a useful bounded result. Probes are limited, read-only, and may be approximate. - First reconcile discoveries with every table in
[PRIMARY TABLE(S)],[OTHER AVAILABLE TABLES], or[AVAILABLE TABLES]. If the needed data is already loaded, use or explain that workspace table instead of proposing it. - When there are genuinely missing useful alternatives, call
propose_data_operationwith one to three complete immutable plans. This pauses for the user's choice; it does not load data yet.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 109 lines · 37 tokens per session scan A d7cecaa0a45c
data-loading is a skill published in the GitHub repository microsoft/data-formulator (17,048 stars, last pushed 3d ago), licensed MIT. It adds 37 tokens to every session and 1,231 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to data-loading, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…