Verify figure caption numerics against their summary.json sidecars. For each figure section in an INDEX.md, walks the linked stats dump, extracts numerics from the caption prose, and matches them. Catches stale prose where a value drifted between figure regeneration and caption update. Sibling to citation-audit…
Review research artifacts for scientific rigor before they're committed or pushed. Catches framing errors (nulls as detection limits, p-values without effect+CI, undefined metrics), unsupported claims, and presentation problems. Sibling to citation-audit (which checks citation structure) and the planned claim-audit…
Local-file research subagent that executes ONE strand against files on disk (the repo / working directory) instead of the web. Uses Read, Grep, Glob to gather evidence and returns it in the per-strand shard format for the orchestrator to persist. The orchestrator dispatches up to 3 of these when local sources are…
Web research subagent that executes exactly ONE research strand. Searches and fetches web sources, extracts atomic claims, scores each source's quality, and writes its OWN two evidence files in the run directory. The orchestrator dispatches one per independent strand, many in parallel.
Report-writer subagent (final stage). Reads ALL per-strand evidence files in a run directory and writes a well-structured, fully-cited report.md. It reads everything from disk so it is robust to context compaction. The orchestrator invokes it once, after retrieval is complete.
Vahtian's company voice + SEO agent. Rewrites and reviews page titles, meta descriptions, and headings so they read in the Vahtian voice AND meet the SEO-compliant company standard (docs/seo-copy-standard.md). Use when auditing or improving titles/descriptions/H1s across vahtian.com, standardising SERP copy before…