feat-dev

A staged workflow for adding a substantial feature to Qwen Code, from investigating the request through design, implementation, testing, review, and revision.

In plain words
What is it for?
It produces a design document and end-to-end test plan, implements the feature, runs verification, audits the result, and incorporates code-review findings.
Why use it?
It gives complex changes a sequence of checks so requirements, existing behavior, edge cases, and integration details are considered before release.

Skill for Claude CodeCodex

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 skills/qwenlm/qwen-code/feat-dev
Any agent
npx skills add QwenLM/qwen-code --skill feat-dev
Clone the repo
git clone --depth 1 https://github.com/QwenLM/qwen-code

Made for: Claude Code, Codex.

Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,076 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 $0.00048 $0.01076
Opus 5 $0.00024 $0.00538
Sonnet 5 $0.00010 $0.00215
Haiku 4.5 $0.00005 $0.00108

Measured 3d ago against content hash 04c0e9df733e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

feat-dev 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 3d ago.

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.

.qwen/skills/feat-dev/SKILL.md · 147 lines

How it starts

The opening of the file, as written. The whole thing — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Feature Development Workflow

Use this workflow when implementing a feature in qwen-code that needs design, behavioral validation, or coordinated changes across multiple files. Each phase produces a concrete artifact. Do not combine phases; the output of each phase feeds the next.

Artifact Paths

Use these paths for planning artifacts:

  • docs/design/<feature>.md
  • .qwen/e2e-tests/<feature>.md

Phase 1: Investigate

Understand the requested behavior and the current qwen-code implementation.

Use a code exploration agent when available. Ask it to inspect the relevant qwen-code areas for:

  • Existing feature definitions: tools, parameters, schemas, commands, UI, or config.
  • Runtime wiring: spawning, lifecycle, state, permissions, hooks, and cleanup.
  • Edge cases and error handling.
  • Integration points and limitations.

In parallel, inspect docs, issues, tests, and nearby implementations that define or constrain the expected behavior. If no exploration agent is available, do the same investigation locally.

Output: mental model of current behavior, desired behavior, constraints, and key file paths with line numbers.

Phase 2: Design Doc

Write a design doc covering:

  • Problem statement and current state, including the behavior gap.
  • Proposed changes by layer or component.
  • Key design decisions and rationale.
  • Files affected.
  • Scope boundaries.
  • Open questions.

Use prose, tables, and bullets. Avoid code snippets unless essential for a key data structure. JSON config examples are acceptable.

Output: design doc on disk.

Phase 3: Test Plan

Use the e2e-testing skill to choose test modes. Then write an E2E test plan covering:

  • Test groups by capability: parameter acceptance, core behavior, error handling, cleanup, and regressions.
  • Exact commands and expected behavior before and after implementation.
  • Unique tmux session names and temp dirs for independent groups.
  • Which groups can be run in parallel by separate test-engineer agents.

Read the full file on GitHub · 147 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. 3d ago First seen · 147 lines · 48 tokens per session scan A 04c0e9df733e

Subscribe to this mod's changes

feat-dev is a skill published in the GitHub repository QwenLM/qwen-code (27,559 stars, last pushed today), licensed Apache-2.0. It adds 48 tokens to every session and 1,076 once invoked, about $0.0002 per session on Opus 5. 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 2026-08-30.