Rek Verifier — Research agent for Claude Code
REK worker (verification).
How to install Rek Verifier
Installs to ~/.claude/agents/contiloop-rek-rek-verifier.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/contiloop/rek/HEAD/agents/rek-verifier.md -o ~/.claude/agents/contiloop-rek-rek-verifier.md Restart Claude Code, or start a new session, for it to be picked up.
What Rek Verifier does
name: rek-verifier description: REK worker (verification). Reads units.md, sources.md, claims.md, and post.md in runs//, plus research.md, post.md, log.md, and profile_hash.txt in repro//, and profiles//research-method.md. Counts how many of the original's development units, rare sources, and base sources the reproduction recovered, and writes runs//repro//verification.md (a yaml block at the very top, then the correspondence tables and the failure routing). It does not
Alternatives in Research
- Deep Research Patcher — Applies critic findings to a nav-deep-research draft as surgical Edit hunks and records a patch log 232 ★
- Verify Figures — A verification agent that performs Stage 1 figure verification — comparing raw figure outputs (PDFs, PNGs from 134 ★
- Fire Fact Checker — Adversarial verification agent that independently attempts to disprove research findings 77 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Rek Method Analyst
REK worker. Extracts the research procedure from one author's traced runs (claims, sources, research-path, uni
Rek Source Tracer
REK worker. Traces every claim in one post back to its source and reconstructs the author's research path. Ful
Profile Synthesizer
Synthesizes voice profiles from self-report scores and computational writing analysis. Merges dual outputs usi
Voice Analyzer
Deep analysis agent for generating voice profiles from writing corpora. Use when analyzing writing samples to
Ms Verifier
Verifies phase goal achievement through goal-backward analysis. Checks codebase delivers what phase promised,
Gsd Verifier
Verifies phase goal achievement through goal-backward analysis. Checks codebase delivers what phase promised,
Related Skills
Jiahao
Dual-profile prompt-as-mental-model harness for LLM agents — advisory generator profile in the working agent,
Doublecheck
Intelligently verify completion claims using parallel completion-verifier agents with smart task splitting
Cmd Judge
Adversarial verifier for the PROVE phase: treats a finished-work report as untrusted claims, re-runs every ver