Flux Review — Development skill for Claude Code
Multi-track deep review — agents across semantic distance (adjacent→orthogonal→distant→esoteric), parallel flux-drive + cross-track convergence.
How to install Flux Review
Installs to ~/.claude/skills/mistakeknot-interflux-flux-review/SKILL.md
mkdir -p ~/.claude/skills/mistakeknot-interflux-flux-review && curl -fsSL https://raw.githubusercontent.com/mistakeknot/interflux/HEAD/commands/flux-review.md -o ~/.claude/skills/mistakeknot-interflux-flux-review/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Flux Review does
name: flux-review description: "Multi-track deep review — agents across semantic distance (adjacent→orthogonal→distant→esoteric), parallel flux-drive + cross-track convergence." user-invocable: true codex-aliases: [flux-review] argument-hint: " [--tracks=auto|2|3|4] [--creative] [--quality=balanced|economy|max] [--interactive]" disable-model-invocation: true
Use the `interflux:flux-review-engine` skill to run a multi-track deep review of the target. Pass `$A
Alternatives in Development
- Mempalace Search — Search your memories across the MemPalace using semantic search with wing/room filtering 58.6k ★
- Research Codebase — You are tasked with conducting comprehensive research across the codebase to answer user questions by spawning 10k ★
- Native Feel Skill — An Agent Skill for designing cross-platform desktop apps that feel native — distilled from Raycast's 2.0 deep 1.9k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Flux Explore
Autonomous multi-round semantic exploration — agents from progressively distant knowledge domains, synthesize
Fetch Findings
Fetch peer findings from a flux-drive review — inspect what agents have shared during a parallel review.
Deep Solve
Solve a hard self-contained problem via the deep-solve convergence harness
Desktop Touch MCP
Windows computer-use MCP server: drive any desktop app via semantic discover-then-act targeting (entities + le
Claude Code Headless
Programmatically drive the real Claude Code CLI in a PTY — structured semantic/screen/committed events, live t
Deep Recall
Deep semantic search over the ENTIRE vault (2k+ notes incl. session-log, handoffs, transcripts, graph) via the
Related Agents
Convergence Monitor
Cross-iteration convergence tracker — runs 6 detection algorithms (score-based, semantic, diminishing returns,
Toy Researcher
Researches ONE toy angle for a specific kid — deep, review-backed, obsession-matched — and returns structured
Deep Repo Auditor
Cross-validated deep repository audit — dual AI models (Sonnet + Codex) in parallel, consolidated report with