A command that guides a feature from its initial idea through implementation, testing, pull request, and merge. A pull request is a proposed code change reviewed before it enters the main branch.

Use it to break work into small batches, write tests first, implement the smallest useful change, run checks at several levels, and prepare a reviewable pull request.

It gives the work a defined completion point and checks that the code, tests, documentation, and repository history agree with the request.

How developerz-ai/universal-lsp configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to developerz-ai/universal-lsp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/developerz-ai/universal-lsp/main/.claude/commands/feature.md

Clone the repo

git clone --depth 1 https://github.com/developerz-ai/universal-lsp

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/commands/developerz-ai/universal-lsp/feature/github.svg)](https://agentmods.dev/commands/developerz-ai/universal-lsp/feature)

Your own site

<a href="https://agentmods.dev/commands/developerz-ai/universal-lsp/feature"><img src="https://agentmods.dev/badge/commands/developerz-ai/universal-lsp/feature/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/commands/developerz-ai/universal-lsp/feature"><img src="https://agentmods.dev/badge/commands/developerz-ai/universal-lsp/feature.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 499 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00017	$0.00499
Opus 5.5	$0.00007	$0.00200
Sonnet 5.5	$0.00003	$0.00100
Haiku 4.5	$0.00002	$0.00050

Measured 2mo ago against content hash 1f34cf1b80fb, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

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 2mo 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/feature.md · 45 lines

What it actually says

/feature

Done means merged and verified. A green local gate is not done; an open PR is not done.

Goal

$ARGUMENTS

Steps

Restate the goal in one line. If two readings exist, say so before building.
Distrust the paperwork. If working from a plan or a doc, check it against the code and git log first — trackers rot in both directions. State plainly which claims you falsified and fix the doc in the same change.
Explore in parallel, read-only, into a ranked worklist of PR-sized batches.
Test first. Bug → reproducing test. Feature → the assertion that defines done. Register suites with ulsp_add_test, named <subsystem>_*.
Build the smallest correct thing. One primitive with its first real caller; no abstraction before a second consumer exists.
Verify by tier: bin/check --files <paths> after each edit, bin/check --changed when the slice compiles, bin/check once before pushing. Fail one, fix one — re-run only the failing command.
PR: <type>/<slug> branch, conventional commit, Closes #N, one concern, ~110 files max. Never push red; wait for checks to register.
Merge queue depth 1. Never two PRs in flight — each merge rebases main under the other and CI results go stale silently.

Fan-out

One checkout. No git worktrees — work lands in a directory nobody is looking at, and each one needs its own configure and build tree. Brief every agent with its exclusive paths, concurrency 1, no git operations, never git stash.

Stop for

An irreversible action, a change to a public contract (the LSP surface, MCP tool names, graph-lite.h, the persistence format), or anything that would need a guard disabled. Those are decisions, not steps.

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.

2mo ago First seen · 45 lines · 17 tokens per session scan A 1f34cf1b80fb

Subscribe to this mod's changes

feature is a command published in the GitHub repository developerz-ai/universal-lsp (0 stars, last pushed yesterday), licensed Apache-2.0. It adds 17 tokens to every session and 499 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-08-31.

C and C++Git and workflow Planning ai-agents cmake code-intelligence code-navigation cpp20

universal-lsp: Command for Claude Code