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/fixgit 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 | $0.00034 | $0.00380 |
| Opus 5 | $0.00017 | $0.00190 |
| Sonnet 5 | $0.00007 | $0.00076 |
| Haiku 4.5 | $0.00003 | $0.00038 |
Grade A, and why
fix 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:fix
Apply one finding's fix from DESIGN.md.critique.md.
Flow
-
Read the critique: open
DESIGN.md.critique.md, find the finding by id. If not found, refuse. If alreadystatus: applied, refuse. -
Show the proposed change: print the finding and the proposed edit to the user.
-
HITL gate: the PreToolUse hook intercepts the next
Edit(named checkpoint:before_design_md_fix) and waits for approval. -
Apply: make the single edit on
DESIGN.md(or on the named source file when the fix touches code). -
Re-lint:
Bash: npx --yes @google/design.md lint DESIGN.md.- If lint regresses, revert the edit and surface the error.
- If lint passes, proceed.
-
Update critique: mark the finding
status: appliedinDESIGN.md.critique.md. -
Append to progress.md: "Applied F. DESIGN.md still passes lint."
Arguments
<finding-id>— required. Must match a finding inDESIGN.md.critique.md.
Boundaries
- One finding at a time. Never batch-applies.
- Always HITL gated.
- Always re-lints. Never leaves DESIGN.md in a regressed state.
See skills/design-harvest/SKILL.md for the full logic.
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 · 40 lines · 34 tokens per session scan A 70836afab86f
fix is a command published in the GitHub repository OthmanAdi/MDDesign (13 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 34 tokens to every session and 380 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
work
通过 v3 流程发起正式工作请求 (entry gate).
tdd
Guide Test-Driven Development workflow (Red-Green-Refactor).
verify
Run comprehensive pre-PR verification (5-phase quality check).
gha
Analyze GitHub Actions CI failures.
commit-push-pr
Create git commit, push to remote, and create/merge pull request.
handoff
Generate context handoff document for session continuity.