Skip to content
Back to skills

Refine

ASecurity

Markilo development — turn the concept the user chose (or a combination of two) into a professional identity (letterforms, spacing, optical balance, icon, small version, lockups, colour), verify it, get one independent critique, and finalize it on the user's approval. Use when the user picks a concept or asks for changes to one.

  • 4 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 7, 2026
developmentgoshellbashnode

Works with

  • claude code

Security analysis

A100/100

Scanned November 7, 2026

npx -y skills add cihantopac/markilo --skill refine --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Refine?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Refine
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/cihantopac-refine/badge)](https://www.skillsdirectory.com/skills/cihantopac-refine)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---

name: refine
description: Markilo development — turn the concept the user chose (or a combination of two) into a professional identity (letterforms, spacing, optical balance, icon, small version, lockups, colour), verify it, get one independent critique, and finalize it on the user's approval. Use when the user picks a concept or asks for changes to one.
argument-hint: "[brand slug, concept id and the feedback]"
allowed-tools: Bash(node "${CLAUDE_PLUGIN_ROOT}/dist/markilo.mjs" *) Bash(node ${CLAUDE_PLUGIN_ROOT}/dist/markilo.mjs *) Read Write Edit Glob Grep
---

# Markilo · Develop, verify and finalize

Request: $ARGUMENTS

Toolchain: `node "${CLAUDE_PLUGIN_ROOT}/dist/markilo.mjs" <command>` (written `markilo`). Run every toolchain command as its own Bash call, written out in full (`node "${CLAUDE_PLUGIN_ROOT}/dist/markilo.mjs" <command> …`). Do not wrap it in shell variables, `cd`, `&&` chains or pipes: the plugin pre-approves exactly that form, so anything else triggers permission prompts. Craft rules: `${CLAUDE_PLUGIN_ROOT}/skills/design/references/design-principles.md`, `${CLAUDE_PLUGIN_ROOT}/skills/design/references/svg-authoring.md`.

1. Collect the user's decision (`markilo feedback --brand <slug>`, or their words) and log it with `markilo answers`.
   Never develop a concept the user dropped. `markilo cost --brand <slug>` before spending.
2. If the user selected without changes and `markilo proposals --brand <slug> --concepts <id>` lists nothing missing
   and `verify` has no failing check, ask whether to finish as chosen; on yes, skip to step 5 without a designer or
   critic. A small, precise change they describe is done directly with the toolchain (`construct`, `lockup`,
   `verify`). Otherwise: one `markilo:concept-designer` agent in `develop <concept id>` mode with the user's changes, pinned spots, or the
   combination (it creates a new concept for a combination). Record `--phase refine` with its usage.
3. `markilo verify --brand <slug> --concept <id>` until nothing fails; `Read` the review, wordmark and context sheets. When the `letters` check says it did not run (a freeform or hand-edited wordmark), read the 32 px wordmark render letter by letter and write down what you read before the brand name; a split or misread letter is fixed before the critic. When `balance` says the mark leans, `markilo center in.svg -o out.svg` shifts it to the optical centre (never run on a locked brand's masters).
4. After a development pass only, one `markilo:design-critic` agent in `final <concept id>` mode. Apply the fixes that improve the work without
   changing what the user chose — once, then verify again; do not critique an unchanged image twice. Record
   `--phase verify`. At the end of each phase, record the session total Claude Code shows:
   `markilo cost --brand <slug> --phase <p> --session-usd <total>` (the total covers the agents; Markilo subtracts what
   they recorded).
5. Show the user the refined identity (`markilo proposals --brand <slug> --concepts <id>`, the gallery), the measured
   checks, the originality verdict and the critic's notes labelled as judgment. Further changes go back to step 2.
6. When the user approves: `markilo finalize --brand <slug> --concept <id> --user-choice "<their words>" --tagline "<tagline>"`
   (`--accept-risk` only on their explicit yes to an `insufficient-evidence` or `needs-review` verdict; add
   `--wordmark-only` when they want the word alone: no symbol lockups, the symbol only makes the app icon and favicon), append the
   decision to `decisions.md`, then offer the export.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.