~ / rigs / jflournoy / claude-analysis-config

jflournoy/claude-analysis-config

Claude Code reviewer agents, guides and a tested dispatch hook for reproducible statistical analysis in R and Stan

↗ GitHub ★ 1 MIT updated 13d ago personal setup Claude Code
stack: JavaScript
share on X
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
CONTEXT TAX · EVERY TURN
~2.8k tokens
Moderate · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details
Quality 0/100 Guardrails v2

Copy this rig

# review before running: this installs third-party code
$ npx degit jflournoy/claude-analysis-config/agents ./rig-claude-analysis-config/agents
$ npx degit jflournoy/claude-analysis-config/commands ./rig-claude-analysis-config/commands
$ npx degit jflournoy/claude-analysis-config/skills ./rig-claude-analysis-config/skills
$ npx degit jflournoy/claude-analysis-config/hooks ./rig-claude-analysis-config/hooks

MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.

$ npx degit jflournoy/claude-analysis-config/skills/research-report .claude/skills/research-report
$ curl -fsSL --create-dirs -o .claude/agents/agent-auditor.md https://raw.githubusercontent.com/jflournoy/claude-analysis-config/main/agents/agent-auditor.md
$ curl -fsSL --create-dirs -o .claude/agents/determinism-reviewer.md https://raw.githubusercontent.com/jflournoy/claude-analysis-config/main/agents/determinism-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/next-priorities.md https://raw.githubusercontent.com/jflournoy/claude-analysis-config/main/agents/next-priorities.md
$ curl -fsSL --create-dirs -o .claude/agents/r-analysis-reviewer.md https://raw.githubusercontent.com/jflournoy/claude-analysis-config/main/agents/r-analysis-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/stan-reviewer.md https://raw.githubusercontent.com/jflournoy/claude-analysis-config/main/agents/stan-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/statistical-analysis-reviewer.md https://raw.githubusercontent.com/jflournoy/claude-analysis-config/main/agents/statistical-analysis-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/usage-estimator.md https://raw.githubusercontent.com/jflournoy/claude-analysis-config/main/agents/usage-estimator.md
$ curl -fsSL --create-dirs -o .claude/agents/voice-authenticator.md https://raw.githubusercontent.com/jflournoy/claude-analysis-config/main/agents/voice-authenticator.md

This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:

{
  "permissions": {
    "deny": [
      "Read(./.env)",
      "Read(./.env.*)",
      "Read(**/.env)",
      "Read(~/.ssh/**)",
      "Bash(git push --force:*)",
      "Read(**/*.pem)",
      "Read(.env)",
      "Read(~/.aws/**)",
      "Read(**/*.key)",
      "Read(**/.env.*)",
      "Bash(git push --force*)",
      "Bash(rm -rf:*)",
      "Bash(rm -rf *)",
      "Bash(sudo *)",
      "Bash(sudo:*)",
      "Bash(rm -rf /)",
      "Bash(git push -f:*)",
      "Read(.env.*)",
      "Bash(git reset --hard*)",
      "Bash(rm -rf /*)"
    ],
    "ask": [
      "Bash(git push:*)",
      "Bash(git push *)",
      "Bash(gh pr merge *)",
      "Bash(git commit:*)",
      "Bash(rm *)",
      "Bash(gh pr merge:*)",
      "Bash(gh pr create *)",
      "Bash(git reset --hard:*)",
      "Bash(npm publish:*)",
      "Bash(git rebase *)"
    ]
  }
}

Skills (1)

Subagents (8)

agent-auditorIntelligently audits other agents for quality, correctness, and relevance
determinism-reviewer
model: sonnet
Finds work being done by model reasoning at runtime that tested code could do reliably. Use when auditing a .claude directory, agents, skills, hooks, or analysis pipelines. Makes recommendations only
next-prioritiesAnalyzes project state to recommend next development priorities and actions
r-analysis-reviewer
model: sonnet
Reviews R analysis code, Rmd and qmd files for silent wrong results — joins that drop rows, type coercion, non-deterministic output, and conclusions that outrun the data. Fires automatically on .R/.Rm
stan-reviewer
model: sonnet
Reviews Stan model code for silent-wrong-answer bugs, geometry problems, and wasted cycles. Fires automatically on .stan edits. Reports findings; never edits the model.
statistical-analysis-reviewer
model: opus
Skeptical peer review of a completed analysis — design, assumptions, inference, and whether the stated conclusion is actually supported. Use on demand before a result is shared or published. Reports f
usage-estimatorAnalyzes project context to provide intelligent Claude usage estimates for development tasks
voice-authenticator
model: opus
Checks reader-facing prose against the voice profile in ~/.claude/guides/voice.md — register, sentence rhythm, em-dash and hedge density, AI tells. Use on report text, figure captions, READMEs, and ab

Slash commands (6)

/commit/hygiene/next/push/refactor-verified/refactor

Similar rigs

copied ✓