editorial-review

Generate a self-contained HTML review artifact for post-Gate-1 investigation findings and feedback export.

43|8|Updated Mar 25, 2026
One-click install
npx skills add https://github.com/buriedsignals/spotlight --skill editorial-review-buriedsignals
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: editorial-review
Source: https://github.com/buriedsignals/spotlight/tree/main/skills/editorial-review
Command: npx skills add https://github.com/buriedsignals/spotlight --skill editorial-review-buriedsignals

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Journalists and investigators need a way to inspect structured findings, verdicts, and source provenance after an OSINT research cycle without running a server or installing tools. This Skill packages case data into a single offline HTML file that renders findings, fact-check verdicts, grounding details, and provenance state, and lets the reviewer export structured feedback as JSON. ## Core Features & Use Cases - Self-Contained Review Artifact: Builds one HTML file with inline CSS and JavaScript that works offline in any modern browser, with no server or CDN dependencies. - Structured Feedback Export: Per-finding and per-source-expression feedback forms serialize into a downloadable review-feedback.json validated against a documented schema. - Provenance and Grounding Display: Renders C2PA signing status, source hashes, evidence references, support types, confidence caps, and expression lifecycle chains. - Use Case: After an investigation's Gate 1 checkpoint, generate review.html from the case directory so an editor can inspect verified findings, flag attribution errors on specific source expressions, and return feedback for a targeted follow-up cycle. ## Quick Start Generate the editorial review artifact for the active case by reading its findings, fact-check, and source-expression data and writing review.html into the case directory.

Frequently Asked Questions about editorial-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create an offline HTML review of investigation findings?

Invoke the skill after Gate 1 with the case directory. It reads findings.json, fact-check.json, and optional source-expression and provenance files, injects them into a self-contained template, and writes review.html that opens in any browser without a server.

How do reviewers export feedback from the review HTML?

The artifact includes per-finding and per-expression feedback forms plus general feedback fields. Clicking the download button serializes all entries into a review-feedback.json file via a Blob download, which is returned to the Gate 1 owner for validation.

Does the review artifact require a server or internet connection?

No. The HTML is fully self-contained with inline CSS and JavaScript, no CDN or external assets. It works offline in recent Chrome, Firefox, Safari, and embedded browsers, and generation is local-only so it works in sensitive mode.

What happens to expression feedback submitted through the review?

Expression feedback is an annotation, not a verdict change. The Gate 1 owner validates expression and finding IDs against current case data, converts items to bounded instructions, and only the independent fact-checker may change a verdict after re-checking.

What are the limitations of the feedback export workflow?

The exported file is not a phase trigger and is never auto-discovered by scanning directories. Feedback text is untrusted free-form input that must be validated and summarized by the Gate 1 owner, and dangling expression or finding references are omitted rather than retargeted.