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.
git clone --depth 1 https://github.com/opensesh/canvas-to-codeWrote 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.
[](https://agentmods.dev/commands/opensesh/canvas-to-code/start)<a href="https://agentmods.dev/commands/opensesh/canvas-to-code/start"><img src="https://agentmods.dev/badge/commands/opensesh/canvas-to-code/start/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.
<a href="https://agentmods.dev/commands/opensesh/canvas-to-code/start"><img src="https://agentmods.dev/badge/commands/opensesh/canvas-to-code/start.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00021 | $0.01558 |
| Opus 5.5 | $0.00008 | $0.00623 |
| Sonnet 5.5 | $0.00004 | $0.00312 |
| Haiku 4.5 | $0.00002 | $0.00156 |
Grade A, and why
start 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/canvas-to-code:start — Universal workflow entry
One command, all eleven gates. Spawns @canvas-to-code-pm.md, which inspects .claude-design/ and .canvas-to-code/state/, then routes by flag or by current phase.
How it dispatches
| Flags | Behavior |
|---|---|
--prep <feature> (or --feature <name> --prep) |
Scaffold only. Creates .claude-design/<feature>/ + screenshots/, source-meta.yaml stub, notes.md stub. No conversation. Refuses if the folder already exists. |
--pr <num> |
Review a slice PR. Spawns @canvas-to-code-reviewer.md with the PR diff + the slice spec from status.json.slices[]. Returns PASS/REVISE with file:line citations. |
--gate <n> |
Jump to gate n (0–10). Re-runs that gate fresh, or for the first time if not yet reached. --gate 5 is the canonical "re-validate the component mapping" shortcut. |
--feature <name> (alone) |
Resolve to that feature and auto-advance from its next pending gate. |
| (no flags) | Guided discovery. PM scans .claude-design/ and presents a menu of what's there (see below). |
Guided discovery (default path)
When called with no flags, the PM agent presents one unified menu so the user picks where the source comes from. PM never silently advances past Gate 0.
-
Scan
.claude-design/(depth ≤ 5). Categorize:- Active features — subfolder has
status.jsonwithphase ≠ "done". - Iter folders (v2) —
iter-*directories withsource-meta.yamlcontainingmetaVersion: 2. The producer (e.g. thepaper-designskill) emits these; the bridge auto-fills feature, subpage, route, and source from the iter'ssource-meta.yaml. - Loose materials — subfolder with
review.html,screenshots/, orsource-meta.yamlbut nostatus.jsonand not an iter folder. - Completed features —
status.jsonwithphase = "done".
- Active features — subfolder has
-
Print one menu:
Welcome to canvas-to-code. Scanning .claude-design/… Found: Active features 1. <feature> (Gate N in progress, last touched <when>) Iter folders (v2) 2. <feature>/<subpage> <iter-name> (<source>, <when>) Loose materials 3. <subfolder>/ (review.html, no status.json) Completed 4. <feature> (done, <date>) 5. Import from external source (Claude Design, Figma, V0, Lovable, Webflow) 6. Start blank — I'll walk you through capturing materials Reply with a number, or paste a path to an iter folder.
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.
- 1mo ago First seen · 103 lines · 21 tokens per session scan A 8b2ff7a2bfb4
start is a command published in the GitHub repository opensesh/canvas-to-code (7 stars, last pushed 7d ago), licensed MIT. It adds 21 tokens to every session and 1,558 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.
Other commands, from other repositories
premium-landing
Plan, build, review, and refine a premium SaaS landing page one section at a time.
style-guide
Generate a style guide with design tokens, type scale, spacing, components, and Tailwind or CSS config.
design-audit
Audit a UI against eight design dimensions and return scored findings with fixes.
brand-identity
Guide a brand identity build from discovery and strategy through visual identity to guidelines.
design-import
Scaffolds React components from a Claude Design handoff bundle and stops at files on disk: no stories, no tests, no pull request. Use when handed a claude.ai/design URL or a local bundle file; when that same scaffold should carry on through test generation, browser verification and an opened PR, run /ork:design-ship…
design-ship
One-shot pipeline turning a claude.ai/design link into a pull request: scaffold via /ork:design-import, stories and specs via /ork:cover, browser verification via /ork:expect, then open the PR. Use when a design link should come back as a PR with no intermediate steps; if all you need is the components written to…