visual-edit

A handler for batches of visual change requests from a design review tool. A designer selects elements in a live preview, adds comments, and sends the collected requests for implementation.

In plain words
What is it for?
Use it to read pending design-tweak requests, edit the named files for each comment in order, and report what changed for every request item.
Why use it?
It prevents comments in one batch from being overlooked or merged incorrectly, and gives each requested change its own progress and result.

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/kirodotdev/kirocrew/visual-edit
Any agent
npx skills add kirodotdev/KiroCrew --skill visual-edit
Clone the repo
git clone --depth 1 https://github.com/kirodotdev/KiroCrew

Made for: Claude Code, Codex.

Per session 50 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,336 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 $0.00050 $0.02336
Opus 5 $0.00025 $0.01168
Sonnet 5 $0.00010 $0.00467
Haiku 4.5 $0.00005 $0.00234

Measured 2d ago against content hash a12f9c710971, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

visual-edit 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.

src/kiro_crew/apps/builtins/design_tweak/skills/visual-edit/SKILL.md · 205 lines

How it starts

The opening of the file, as written. The whole thing — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Visual Edit — Design Tweak request handling

The Design Tweak app lets a designer visually pick element(s) in a live preview and attach natural-language comments. Comments accumulate into a request, and the designer sends the whole batch at once. One request file therefore contains many comments as sub-items:

~/.kiro/crew/apps/design-tweak/data/queue/<timestamp>-<id>.json

A request is a batch. Work every comment in it, and report per comment.

The loop

  1. Read the request file. The prompt names the request id; if not, read the newest file in the queue dir. state will be "sent" — a "draft" request has not been handed to you yet, so leave drafts alone.
  2. Work comments[] in order. Each entry is an independent edit with its own cid, comment, sourceFile, and selection. Do not merge them into one change, even when two comments touch the same file.
  3. For each comment, resolve the target, make the edit, and report against that comment's cid (see below).
  4. When every comment is done, tell the user what changed, grouped by comment number.

Reporting progress — always per comment

Each comment has its own progress bubble in the preview, so notes must be addressed to a cid:

POST /apps/design-tweak/api/thread?id=<requestId>&cid=<commentId>
body: {"role": "agent", "text": "Editing styles.css — uppercasing .section-title"}

Post one note when you start a comment, and one per meaningful step. Keep each to a single short line. When that comment is finished, post a final note carrying a status so its dot turns green:

POST /apps/design-tweak/api/thread?id=<requestId>&cid=<commentId>
body: {"role": "agent", "text": "Done — added text-transform: uppercase", "status": "done"}

The request's own status is derived: it flips to done on its own once every comment is done. Never set it directly.

Never write to the request file yourself. Do not edit state, status, sentAt, or any other field by writing the JSON — the panel derives the request's badge from its comments, and a hand-written state desynchronises them. Progress is reported only through POST /thread. The file is the app's state, not a scratchpad.

Read the full file on GitHub · 205 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 · 205 lines · 50 tokens per session scan A a12f9c710971

Subscribe to this mod's changes

visual-edit is a skill published in the GitHub repository kirodotdev/KiroCrew (3,532 stars, last pushed today), licensed Apache-2.0. It adds 50 tokens to every session and 2,336 once invoked, about $0.0003 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.