midnight-fact-check

Fact-check content against the Midnight ecosystem. Extracts claims, classifies by domain, verifies each via midnight-verify, and produces a report.

Fast fact-check content against the Midnight ecosystem. Extracts claims, verifies each via source inspection (skipping classification and execution), and produces a report.

Use this agent to extract testable claims from a chunk of Midnight-related content. Dispatched by the /midnight-fact-check:check command in Stage 1, one instance per content chunk, running in parallel. The agent reads its assigned content (file paths provided in the dispatch prompt), identifies all testable claims, and returns them as a JSON array. Example: Dispatched with a skill's SKILL.md and its references/ folder. The agent reads all files, identifies claims like "persistentHash returns Bytes<32>" and "for loops use lower..upper syntax", and returns a JSON array of claim objects.

Use this agent to classify claims by domain. Dispatched by the /midnight-fact-check:check command in Stage 2, one instance per domain (compact, sdk, zkir, witness), running in parallel. Each instance receives its assigned domain and a copy of the claims file. It reads the copy, tags claims belonging to its domain, writes the updated copy, and returns a summary. Example: Dispatched as the "compact" classifier. Reads claims file, tags claims about Compact syntax, types, stdlib, and compiler behavior with domain "compact". Writes updated copy. Returns "Tagged 45 claims as compact."

This skill should be used when classifying fact-check claims into verification domains (Compact, SDK, ZKIR, Witness). Covers how to tag claims with their domain, assign classification confidence, handle cross-domain claims, and resolve boundary cases between domains like compiler behavior vs compiled output. Triggered by queries like "classify these claims", "tag claims by domain", "which domain does this claim belong to", or "run domain classification on the claims file". Used by the domain-classifier agent in the midnight-fact-check pipeline.

This skill should be used when extracting testable claims from Midnight documentation or source content. Covers how to identify verifiable statements about Compact syntax, types, APIs, compiler behavior, and runtime errors. Defines the JSON output schema for structured claim lists. Relevant when asked to "extract claims", "find testable statements", "parse documentation for verifiable facts", or "produce a claim list from content chunks". Used by the claim-extractor agent in the midnight-fact-check pipeline.

This skill should be used when generating fact-check reports, formatting terminal summaries of verification results, or creating GitHub issues for refuted claims. Covers the markdown report template (executive summary, results by domain, refuted claims table), the terminal summary block, and GitHub issue templates (per-claim, per-file, and summary). Triggered by queries like "generate the fact-check report", "format verification results", "create a GitHub issue for refuted claims", or "what does the terminal summary look like". Used in Stage 4 of the midnight-fact-check pipeline to assemble final output from verification results.