dev-loop

A Claude Code workflow made of several read-only code-review and refactoring reviewers, called lenses. It gathers their suggested fixes and passes accepted findings to TDD, meaning tests are written or updated before the code change is implemented.

In plain words
What is it for?
Use it to review a change for correctness, security, maintainability, refactoring opportunities, and test quality, then implement accepted findings under green tests.
Why use it?
It separates finding problems from editing code, while keeping implementation tied to passing tests.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/greendwin/mcp-tasker/dev-loop
Clone the repo
git clone --depth 1 https://github.com/greendwin/mcp-tasker
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 581 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00581
Opus 5 $0.00000 $0.00291
Sonnet 5 $0.00000 $0.00116
Haiku 4.5 $0.00000 $0.00058

Measured yesterday against content hash a19493710e7d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

dev-loop scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

docs/agents/dev-loop.md · 54 lines

How it starts

The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Dev loop

This repo runs inside Claude Code. The code-review lens invokes the built-in /code-review command (read-only by default); the refactor lenses are read-only reviewers that propose refactorings in prose. Every lens reports findings with an inline suggested-fix and never edits the tracked tree — dev-loop hands the accepted findings to tdd, the sole writer, which implements them under green tests. Each lens below is one reviewer; dev-loop spawns the lenses in a roster in parallel and collects their findings. This document is self-contained — no other file is needed to perform any lens below.

code-reviewer

Runs against the implemented change once it is green.

general

Invoke the /code-review built-in command over the change under review. Report every issue it surfaces — correctness, missing/weak tests, security, and maintainability — as findings. Do not edit code; propose only.

tests

Review the change specifically for test quality: every new public behavior has a behavior-level test, tests exercise the public interface (not internals), and there are no implementation-coupled or mock-heavy tests. Report gaps and weak tests as findings. Do not edit code; propose only.

refactor-reviewer

Runs against the whole change during the refactor phase.

duplication

Review the change for repeated logic or structure — copy-pasted blocks, parallel branches that differ only in a value, the same computation done in several places. For each, name the duplicated sites and propose how to unify them (extract a helper, parameterize, hoist a shared value). Report each as a finding with location, rationale, and an inline suggested-fix. Do not edit code; propose only.

thermo-nuclear-code-quality-review

Invoke the thermo-nuclear-code-quality-review skill over the whole change for an extremely strict maintainability and structure audit. Be ambitious: hunt for "code judo" restructurings that preserve behavior while making whole branches, helpers, modes, or layers disappear entirely — prefer deleting complexity over rearranging it. Flag, as distinct findings: any file the change pushes past ~1000 lines without a strong structural reason; ad-hoc conditionals or one-off special cases tangled into unrelated flows (spaghetti growth); thin or pass-through abstractions and identity wrappers that add indirection without buying clarity; and type/boundary smells — needless optionality, any/cast-heavy code, or silent fallbacks papering over an unclear invariant. Report each as a finding with location, rationale, and an inline suggested-fix. Do not edit code; propose only.

Read the full file on GitHub · 54 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 54 lines · 0 tokens per session scan A a19493710e7d

Subscribe to this mod's changes

dev-loop is an agent published in the GitHub repository greendwin/mcp-tasker (0 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 581 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens