Dr Efficiency — Development agent for Claude Code
Deep-review finder angle.
How to install Dr Efficiency
Installs to ~/.claude/agents/demchaav-deep-review-dr-efficiency.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/DemchaAV/deep-review/HEAD/agents/dr-efficiency.md -o ~/.claude/agents/demchaav-deep-review-dr-efficiency.md Restart Claude Code, or start a new session, for it to be picked up.
What Dr Efficiency does
name: dr-efficiency description: Deep-review finder angle. Finds wasted work the diff introduces - repeated expensive construction, re-parsing, uncached lookups, per-item process spawns, whole-tree walks - and measures the cost where it can. tools: Read, Grep, Glob, Bash, Write
You are one finder angle in a multi-angle code review. Your run context - worktree, base, head, diff paths, findings file - arrives in the prompt.
Your angle: efficiency
You are looking for work the program
Alternatives in Development
- Python Sidecar Specialist — Use for the Python sidecars on both surfaces: the verify checks (syntax, imports, statement mapping, coverage 179 ★
- Ds Flash — DSH (DeepSeek Harness) worker on the flash tier - fast and cheap 116 ★
- Book Disruptor — Chaos agent for the book pipeline 114 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Simplify Efficiency Reviewer
Read-only efficiency lens of the HIGH simplify pass: redundant loops, repeated parsing/I-O/queries/allocations
Ccd Reviewer
Reviews code in areas the risk ledger measures as hostile to Opus 5, where an Opus 5 reviewer would be refused
Dr Wrapper Correctness
Deep-review finder angle. Audits every wrapper, adapter, proxy, shim, facade and projection the diff introduce
Typesafe Calibrator
Turns invented thresholds into measured ones. Builds labelled sets, plots confidence against accuracy, measure
Ndv Optimize
Performance optimization specialist. Use when code is slow, queries are expensive, bundle is large, or latency
Treasurer
Efficiency auditor (Sonnet) that checks whether agents drifted - repeated failing approaches, work nobody aske
Related Skills
Noise Budget
Cut wasted token spend without skipping real bugs. Distinguish IMPOSSIBLE work (tech-mismatch CVEs, encoded re
ToolTax
Find wasted tokens, slow tools, duplicate calls, and expensive MCP servers in AI agent workflows.
Claude Code Ledger
Local-first cost & tool-call audit for Claude Code. Finds wasted context: repeated reads, blind retries, huge