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.
npx agentmods add commands/othmanadi/mddesign/injectgit clone --depth 1 https://github.com/OthmanAdi/MDDesignWhat 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 | $0.00031 | $0.00282 |
| Opus 5 | $0.00015 | $0.00141 |
| Sonnet 5 | $0.00006 | $0.00056 |
| Haiku 4.5 | $0.00003 | $0.00028 |
Grade A, and why
inject 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.
What it actually says
/mddesign:inject
Manually run the design-bridge skill's token-inject logic.
What I do
GlobDESIGN.mdat project root. If absent, surface/mddesign:harvestnudge and stop.Readtask_plan.mdto identify the active phase.- Extract the token slice relevant to the active phase (using the same heuristics as the UserPromptSubmit hook).
Editfindings.mdto add or replace the## Design Contextblock with the current slice.
Full logic in skills/design-bridge/SKILL.md.
Why use this instead of waiting for the hook
The UserPromptSubmit hook only fires on UI-keyword detection. If you know you are about to start a UI task but your prompt does not have obvious UI keywords, /mddesign:inject gets the context into findings.md preemptively.
Boundaries
- Never edits
DESIGN.mdortask_plan.md. - Only edits
findings.mdunder the## Design Contextsection.
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.
- 2d ago First seen · 27 lines · 31 tokens per session scan A 02b7fe0a9012
inject is a command published in the GitHub repository OthmanAdi/MDDesign (13 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 31 tokens to every session and 282 once invoked, about $0.0002 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.
Other commands, from other repositories
design-audit
/design-audit [屏幕名称] /design-audit --full # 全应用审计 /design-audit login # 审计登录页面 /design-audit dashboard # 审计仪表盘.
work
通过 v3 流程发起正式工作请求 (entry gate).
accessibility
You MUST start your FIRST response with this exact agent activation line.
verify
Run comprehensive pre-PR verification (5-phase quality check).
commit-push-pr
Create git commit, push to remote, and create/merge pull request.
fix
Quick fix for build, lint, and type errors.