ds-help
Where the migration stands, what blocks it, and the next command.
Skill by renzobianchi
Chalk Labs
Need help deploying ds-help?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/renzobianchi-ds-help
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/ds-help
What's inside
- Skill name
ds-help- Description
- Where the migration stands, what blocks it, and the next command. Run it whenever you are lost.
- License
- MIT
- Instructions
- 61 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- files: referenced but missing: scripts/ds-manifest.mjs
Files (1)
- SKILL.md
SKILL.md (first lines)
Orientation for a design-system migration run with this plugin. Read the repo, never memory: the artifacts below are the only state. Commands are written namespaced (plugin install); a flat install drops the `ds-ai-ready:` prefix. ## 1. Detect the phase Evaluate every row top to bottom; the current phase is the **last** row whose signal holds (rows are cumulative, so a later row holding with an earlier one missing still counts). Every path is relative to the repo root or to `ds.config.json` (`manifests.*`, `namespace`). | Signal | Phase reached | Next command | |---|---|---| | nothing below holds | before foundations | `/ds-ai-ready:ds-migrate foundations` | | `ds.config.json` exists | foundations started | `/ds-ai-ready:ds-migrate foundations` | | `scripts/ds-manifest.mjs` exists and `node scripts/ds-manifest.mjs check` exits 0 | foundations: guards installed | `/ds-ai-ready:ds-migrate foundations` (finish steps 3 to 6) | | a `.tsx` and a `.stories.tsx` under `namespace` | foundations done | existing library: `/ds-ai-ready:ds-migrate triage`; from zero: kit alignment (`phases/kit.md`, design owner) | | `TRIAGE.md` and `manifests.legacy/*.json` exist (skip when `legacyPath` is empty or absent) | triage done | kit alignment | | at least one `manifests.parity/*.json` at `kit-ready` | kit producing | `/ds-ai-ready:ds-migrate component <key>` on the first `kit-ready` by dependency order | | every `manifests.parity/*.json` at `parity` or `code-only`, no `kit-ready`/`kit-wip`/`gap-*` | component queue drained | `/ds-ai-ready:ds-migrate cleanup` (starts with §0, the re-audit) | | `node scripts/ds-manifest.mjs check --kit-version <current>` and `audit code` both exit 0 | re-audit done | `/ds-ai-ready:ds-migrate cleanup` (§1 onward; `1.0.0` unblocked) | | `legacyPath` gone and every legacy manifest `replaced`/`absorbed`/`deprecated` | cleanup done | `/ds-ai-ready:ds-patterns inventory`, then `/ds-ai-ready:ds-migrate steward` | | `governance.model` non-empty in `ds.config.json` | steward done, the system is live | rerun the count in `phases/steward.md` step 1 every quarter | A component in progress: a branch named `<prefix>-<key>` or a manifest whose `note` mentions an open PR. Report it as "in flight" with the step of `phases/component.md` it appears to be at, judged by what exists (branch, stories, manifest at `parity`, PR open, reviewers assigned). ## 1a. Parity health Read the kit's current version id from the bridge (Figma: file version; Paper: project commit), then put every `status: parity` manifest in exactly one bucket, by the first test that holds: - **verified**: `verified.kitVersion` equals the current id and `verified.level` is at or above `parityLevel` in `ds.config.json`; - **stale**: `verified` present with a different `kitVersion`; - **unverified**: no `verified` field, or `verified.level` below `parityLevel`; - **code drift**: the key appears in the output of `node scripts/ds-manifest.mjs audit code`. The procedure for the last three buckets is `../ds-migrate/references/phases/cleanup.md` §0. With the bridge down, read the kit version as unknown, say so in the report, and fill the last three buckets from the files. ## 2. Find what blocks Check each; list only the ones that hold: - `design.figmaFileKey` or `design.kitName` empty with `designTool: figma` (or `paperProjectPath` empty with `paper`): rule 1 forbids building any component. Owner: design. - `node scripts/ds-manifest.mjs check` failing: quote its output verbatim; each line names the module and the fix. - Any stale, unverified or code-drift entry from §1a: quote the `check --kit-version <current>` and `audit code` output verbatim, name cleanup §0 as the fix, and report `1.0.0` as blocked until the three buckets are empty. - A `parity` component without `manifests.usage/<key>.md`: `node scripts/ds-manifest.mjs usage <key>`, then fill it. - `kit-ready` manifests whose composed parts are not yet `parity`: name the part. - `FINDINGS.md` present: count the entries marked **Open** and name the oldest. - A `deprecated` parity manifest without `supersededBy` or `removeIn`: the guard names it; `phases/steward.md` step 4 has the fix. - `manifests.ship` set and `npm pack --dry-run` missing `llms.txt`, `<ship>/usage/*.md`, or (from the first `replaced` legacy export) `<ship>/MIGRATION.md`: a consumer's agent gets compiled JS only. `check` names the missing `files` entry or script; `phases/foundations.md` step 4 has the wiring. ## 3. Report Print, in this order, nothing else: 1. **Phase**: the row from step 1, one line, plus the in-flight component if any. 2. **Parity**: `<n> verified against kit <v> · <n> stale · <n> unverified · <n> code drift`, then the stale and unverified keys on one line when they are ten or fewer together. When any of the three is non-zero, print this glossary line unchanged: `stale = the kit moved after the stamp · unverified = parity claimed, never proven · code drift = the source moved away from its manifest`. 3. **Blocked by**: the list from step 2, or "nothing". 4. **Next**: the single command or hand-off from the table, with the key filled in. 5. **Skills**: `ds-migrate <phase>` (runs one phase end to end, `steward` last), `ds-patterns <step>` (after cleanup), `ds-rules` (fires on its own under the namespace; ask it to state rule 1 to confirm it loaded), `ds-help` (this). 6. **Read next**: the absolute path of the one phase file for the next command, resolved from this skill's folder: `../ds-migrate/references/phases/<phase>.md` (`kit.md` for a design hand-off), or `../ds-patterns/references/patterns.md` after cleanup. Done when every line above is backed by a file or command output you read this turn.
From renzobianchi/skills, MIT.
More from renzobianchi/skills
- 0.600score
arena-design
renzobianchi
Design-exploration arena: N parallel candidates with assigned directions attack the same UX+UI problem in Paper, Figma, or code; visual cross-judge, base +…
SkillAFreshness: A★ 2Claude Code +9 - 0.559score
homepage-messaging
renzobianchi
From a positioning doc to a homepage: value propositions on the Messaging House, section outline, hero choice, element counts; hands finished lines to…
SkillBFreshness: B★ 2Claude Code - 0.559score
gtm-readiness
renzobianchi
Go-to-market audit for a B2B startup: which fit it has (find, sell, serve, retain), which GTM phase it is in, whether the segment is real, and the mistakes in…
SkillBFreshness: B★ 2Claude Code - 0.559score
startup-positioning
renzobianchi
Guided positioning interview for a B2B product you do not own: readiness gate, reconnaissance, four interview rounds, candidate strategies with thesis and…
SkillBFreshness: B★ 2Claude Code - 0.557score
html
renzobianchi
Build standalone interactive HTML artifacts as the medium for agent↔designer conversation — review docs with approve/deny/discuss, N-way prototype variants…
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
ds-rules
renzobianchi
Rules for building components in a shadcn-based design system with a Figma or Paper kit.
SkillBFreshness: B★ 2Claude Code +9 - 0.554score
ds-patterns
renzobianchi
Post-migration phase: inventory reused layouts, name them in the kit, write rules beside them, run the drift test, set up the product knowledge base and…
SkillBFreshness: B★ 2Claude Code +9 - 0.554score
ds-migrate
renzobianchi
Run one phase of a design-system migration to shadcn (AI-ready): triage, foundations, kit, parity, component <key>, cleanup, steward.
SkillBFreshness: B★ 2Claude Code +9
Signals
- Stars
- 2
- Forks
- 2
- Open issues
- 0
- Last commit (this folder)
- Sep 10, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/renzobianchi/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from renzobianchi
- 0.600score
arena-design
renzobianchi
Design-exploration arena: N parallel candidates with assigned directions attack the same UX+UI problem in Paper, Figma, or code; visual cross-judge, base +…
SkillAFreshness: A★ 2Claude Code +9 - 0.559score
homepage-messaging
renzobianchi
From a positioning doc to a homepage: value propositions on the Messaging House, section outline, hero choice, element counts; hands finished lines to…
SkillBFreshness: B★ 2Claude Code - 0.559score
gtm-readiness
renzobianchi
Go-to-market audit for a B2B startup: which fit it has (find, sell, serve, retain), which GTM phase it is in, whether the segment is real, and the mistakes in…
SkillBFreshness: B★ 2Claude Code - 0.559score
startup-positioning
renzobianchi
Guided positioning interview for a B2B product you do not own: readiness gate, reconnaissance, four interview rounds, candidate strategies with thesis and…
SkillBFreshness: B★ 2Claude Code - 0.557score
html
renzobianchi
Build standalone interactive HTML artifacts as the medium for agent↔designer conversation — review docs with approve/deny/discuss, N-way prototype variants…
SkillBFreshness: B★ 2Claude Code +9 - 0.557score
ds-rules
renzobianchi
Rules for building components in a shadcn-based design system with a Figma or Paper kit.
SkillBFreshness: B★ 2Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get ds-help running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
ds-help skill. Agent Mart. https://agentmart.dev/skills/renzobianchi-ds-help (last updated 2026-10-24)