Reviewer Correctness — Data & AI agent for Claude Code
Correctness specialist reviewer.
How to install Reviewer Correctness
Installs to ~/.claude/agents/arcasilesgroup-claudeline-reviewer-correctness.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/arcasilesgroup/claudeline/HEAD/.claude/agents/reviewer-correctness.md -o ~/.claude/agents/arcasilesgroup-claudeline-reviewer-correctness.md Restart Claude Code, or start a new session, for it to be picked up.
What Reviewer Correctness does
name: reviewer-correctness description: "Correctness specialist reviewer. Verifies code actually works as intended: intent-implementation alignment, integration boundary correctness, logic errors, data flow integrity, and behavioral change analysis. Dispatched by ai-review." model: opus color: orange tools: [Read, Glob, Grep, Bash] mirror_family: specialist-agents generated_by: ai-eng sync canonical_source: .claude/agents/reviewer-correctness.md edit_policy: generated-do-not-edit
You a
Alternatives in Data & AI
- Authz Reviewer — PYA specialist spawned by the /pya orchestrator 240 ★
- Review Code Correctness — A code review agent that checks analysis scripts for logical errors, incorrect function usage, wrong variable 134 ★
- PDF Debugger — Debug PDF processing issues — OCR failures, text extraction problems, AI misclassification, and renaming error 116 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Sf Org Verifier
Verifies a Salesforce change in the live org after deployment - smoke probes, data and limit queries, log capt
Read Only Explorer
Maps how code actually works — call paths, call sites, data flow, existing patterns, ownership boundaries, and
Melchior
MELCHIOR-1, the scientist. Judges a proposition on logic, data, and correctness. Returns a strict JSON verdict
Admin Portal Architect
Use when auditing the admin console/back-office: which pages actually work, silent empty results, dead routes,
Lens Instructions
Reviews LLM instruction files (.md prompts, protocols, skill/agent definitions, templates) for cross-file cons
Dx Auditor
Audits the framework by actually using it — scaffolds an app, runs the generators, makes realistic mistakes, a
Related Skills
Check Staged
Inspect the staged changes for correctness, logic integrity, and impact relative to the last commit. Stash uns
Stakeholder Translator
Shows how 4 different executives will actually read your message — not how you intend it, but how each person'
P Reality Check
Assess whether the project actually has the intended thing