Build the query plan from artifacts, not guesses.
This is the offline SQL workflow: no Gateway, no database, and no runner. The AI works from exported Named Queries, schemas, screenshots, Perspective JSON, errors, and pasted rows, then hands back bounded SQL guidance plus the target-side checks still required.
Why it was made
- Uses customer-provided artifacts as the source of truth instead of inventing Named Query paths, database connections, tables, columns, or parameters.
- Keeps dashboard work read-only and bounded by default, with write SQL requiring explicit user direction.
- Turns every offline answer into a target-side verification checklist for Designer, the browser, or the database client.
- Keeps runner/API instructions, local paths, credentials, and test-only resource names out of the no-API skill.
Common use cases
- Reviewing exported Named Query resources.
- Designing bounded SQL from a schema snippet or screenshot.
- Fixing Perspective binding parameter maps from JSON.
- Preparing a customer to test a query in Designer or their database tool.
Proof style
Validate the artifact, not the promise.
The skill is explicit about unverified claims and gives the user a checklist to prove the result in the target environment.
Evidence
Offline SQL workflow proof
The graphic is AI-generated for the site. The checks below come from sanitized no-API SQL evidence: product-boundary scans, packaging smoke checks, and offline wording synchronized from API-backed SQL and browser tests without adding runner workflow to this skill. It prepares target-side tests; it does not claim live Gateway or database execution.
- No-API product scans found no runner action names, Web Dev workflow tokens, local sample query paths, test-run IDs, credentials, or environment-specific paths in the shipped skill.
- The shipped zip contains only SKILL.md, agents/openai.yaml, and the historian guardrails reference; evidence, testing, handoff, and raw notes stay out of the product package.
- Offline guidance was synchronized from API-backed SQL tests for Perspective filters, URL params, Table selection/detail bindings, scanner boundaries, Named Query call shapes, and target-side verification patterns.
- The skill tells the AI to work from user-provided artifacts and label placeholders instead of inventing Gateway project names, Named Query paths, database connections, tables, columns, or parameters.
- Dashboard SQL stays bounded and read-only by default; destructive or write SQL is blocked unless the user explicitly requests it.
Example prompt
Review this exported Named Query and Perspective binding JSON offline for a TextField search table.This is the kind of short instruction the skill is designed to make useful. The value is in the checks and Ignition-specific defaults the AI uses after reading the skill.
Expected output
- Maps the exported params and Perspective binding fields before writing SQL.
- Proposes one bounded read-only SELECT/WITH statement with placeholder names.
- Lists Designer preview, browser state, row-count, empty-state, and hostile-input checks.
Related workflows
Skills that pair with this one
Lifetime membership
Want access to this skill?
The one-time membership unlocks this skill, the full Ignition AI Toolkit, the Web Dev runner materials, setup docs, versioned downloads, and future Ignition 8.3 skills as they ship.
No subscription. Lifetime members keep portal access to the current toolkit and future skill releases, including planned Ignition 8.3-focused skills as they are added.