data-quality skillA
data-quality is agent-read markdown (skill) from nahid-sparktales/agent-dispatcher: Decide what must be true of a dataset and where that assertion belongs — freshness against event time, volume floors and bands, distribution and referential checks, source reconciliation, and whether it should be a schema constraint, a blocking publish gate or an alert. Use when adding or reviewing data checks, when a wrong number reached a consumer and nothing caught it, or when a check fires so often it is being ignored. Not for proving a specific migration did what it claimed (database-migrat.
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Data quality assertions A wrong number looks exactly like a right one. Nothing about a dashboard, a type check or a green job reveals that the load was half-empty, that a join fanned out, or that a column has been null since Tuesday. The only thing that reveals it is an assertion someone wrote in advance — placed where it fires before a consumer acts on the number, not after. ## When this fires Checks are being added to or reviewed on a dataset; a bad value reached a consumer and no check caught it; or an existing check is noisy enough that people are muting it. It does not fire for a one-off verification of a migration, for enforcing an invariant inside a single application write path, or for deciding what a metric means. ## Procedure 1. **Start from the failure that hurt, not from a catalog of check types.** Which wrong value reached whom, how long it took to notice, and what would have caught it earliest. Generic not-null checks generated across four hundred columns produce noise, and a noisy channel is a muted channel — that is a net loss, not a neutral one. 2. **Place each assertion on the lowest rung that can hold it.** This is the load-bearing decision …
Read the whole file at its exact version.
How to install
mdr add nahid-sparktales/agent-dispatcher/data-quality@git:20260919.a0d4f55mdr add nahid-sparktales/agent-dispatcher/data-quality@sha256:38cb5357f2af0cd4Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_i52uw46fveb5cvih)
1 badge views in 30 days
Versions
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (8504 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
nahid-sparktales/agent-dispatcher · 49 stars · license MIT · pushed 2026-09-23 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_i52uw46fveb5cvih GET https://markdownregistry.com/api/v1/resolve?ref=nahid-sparktales/agent-dispatcher/data-quality GET https://markdownregistry.com/api/v1/blob/38cb5357f2af0cd40ed099bb632a5e952dfdbe71eeda6bbce94e9eb8c564fcde
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
More from nahid-sparktales/agent-dispatcher
Every file in nahid-sparktales/agent-dispatcher