next-feature

A command that selects the next eligible feature from a feature list, asks a development skill to build it, validates the result, updates the list, commits the work, and prints a summary.

In plain words
What is it for?
Use it to pick pending features whose dependencies are finished and carry them through implementation, checks, metadata updates, and a Git commit.
Why use it?
It turns a multi-step feature workflow into one repeatable run, reducing the chance of skipping selection, validation, or bookkeeping.

Command for Claude Code

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 commands/liam-machine/erd-studio/next-feature
Clone the repo
git clone --depth 1 https://github.com/liam-machine/erd-studio

Made for: Claude Code.

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 659 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 $0.00659
Opus 5 $0.00000 $0.00329
Sonnet 5 $0.00000 $0.00132
Haiku 4.5 $0.00000 $0.00066

Measured 2d ago against content hash f2370186e570, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

next-feature 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 2d 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.

.claude/commands/next-feature.md · 51 lines

How it starts

The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You MUST complete ALL steps below in a single run — selecting the feature, building it, validating, updating features.json, committing, and printing the summary. Do NOT stop, pause, or ask for input between steps. The job is not done until the final summary is printed.

Step 1 — Select the next feature

  1. Parse features.json and find the first feature where status is "pending", skip is not true, and all dependencies have status: "completed" (in ID order). If $ARGUMENTS contains a feature ID, select that specific feature instead.
  2. If no eligible feature is found, show a progress summary and stop.
  3. Print the feature ID, name, and description.

Step 2 — Build the feature

  1. IMMEDIATELY use the Skill tool to invoke skill feature-dev:feature-dev. You MUST call the Skill tool — do NOT implement the feature yourself and do NOT explore the codebase beforehand. Pass the feature name, description, implementation_notes, and validation_criteria as the args string.

Step 3 — Validate (after feature-dev returns)

The feature-dev skill only writes code. When it returns control to you, you MUST immediately continue with validation — do NOT stop, summarise, or wait for user input.

  1. Run the following checks in order. Fix any failures before proceeding.

    a. Type-checknpm run compile (tsc --noEmit for both tsconfigs). Fix type errors.

    b. Buildnpm run build. Verify both bundles produced without errors.

    c. Unit testsnpm test. Verify all tests pass.

    d. Extension Development Host — If any validation criteria begin with "Verify in Extension Development Host:", launch VS Code with code --extensionDevelopmentPath="$(pwd)" and use Chrome browser automation (claude-in-chrome) to verify UI-visible criteria. Skip if the feature is purely backend with no UI criteria.

Step 4 — Update features.json

  1. Update the feature's status from "pending" to "completed" in features.json.
  2. If all features in the feature's phase now have status: "completed", also update that phase's status to "completed".

Read the full file on GitHub · 51 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. 2d ago First seen · 51 lines · 0 tokens per session scan A f2370186e570

Subscribe to this mod's changes

next-feature is a command published in the GitHub repository liam-machine/erd-studio (5 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 659 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.