Borrowing it
Nothing to install: this file belongs to volomydyr/design-engineer-plugin. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/volomydyr/design-engineer-plugin/main/CLAUDE.mdgit clone --depth 1 https://github.com/volomydyr/design-engineer-pluginWrote 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/instructions/volomydyr/design-engineer-plugin/claude-md)<a href="https://agentmods.dev/instructions/volomydyr/design-engineer-plugin/claude-md"><img src="https://agentmods.dev/badge/instructions/volomydyr/design-engineer-plugin/claude-md/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/instructions/volomydyr/design-engineer-plugin/claude-md"><img src="https://agentmods.dev/badge/instructions/volomydyr/design-engineer-plugin/claude-md.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 | $0.17303 | $0.17303 |
| Opus 5.5 | $0.06921 | $0.06921 |
| Sonnet 5.5 | $0.03461 | $0.03461 |
| Haiku 4.5 | $0.01730 | $0.01730 |
Grade D, and why
design-engineer-plugin CLAUDE.md scanned grade D with 2 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 2mo 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
**Debugging**: run `tail -f ~/.claude/cache/de-process-recall.log` to verify the hook fires when expected. Each fire appends a single line of the form `[ISO_TIMESTAMP] FIRED | workflow=<name> cwd=<path>`. Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
The default `<YYYY-MM-DD-HHMMSS>` timestamp pattern for scratch captures keeps debug output organized by session — easy to scan, easy to delete a day's worth in one `rm -rf .design-engineer-plugin/temporary/playwright/20 How it starts
The opening of the file, as written. The whole thing — 945 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design Engineer Plugin Development
Skill loading from commands (doc-compliant pattern)
This plugin's skills all set disable-model-invocation: true (intentional — skills are libraries loaded by commands, not auto-discoverable workflows). The Skill tool will REJECT any attempt to invoke them programmatically with the error: Skill <name> cannot be used with Skill tool due to disable-model-invocation.
Therefore, never tell the model to "load the X skill" or "invoke the Y skill" in a command body. The model interprets that as a Skill-tool call and the command crashes on the very first run.
The required pattern
Every command file MUST start its body with this short note (placed just after the # Title heading and before any other section):
## Plugin paths
Your conversation context contains a line `DESIGN_ENGINEER_PLUGIN_ROOT: <absolute path>` injected by the plugin's UserPromptSubmit hook. Whenever this command references `${DESIGN_ENGINEER_PLUGIN_ROOT}/...`, substitute the absolute path from that context line. No shell commands are run from this command body.
This relies on a single, permission-free mechanism: the plugin's UserPromptSubmit hook (hooks/de-start-state.sh) injects DESIGN_ENGINEER_PLUGIN_ROOT: <abs path> as additionalContext text on every prompt. The model sees that line in its context and uses it as the substitution value when it encounters ${DESIGN_ENGINEER_PLUGIN_ROOT}/... in the command body.
Mechanisms NOT to use, and why
- Bash injection (
!`...`) — documented at https://code.claude.com/docs/en/slash-commands.md#inject-dynamic-context, but Claude Code's permission system blocks!-prefix patterns at command-load time in Auto mode and any restrictive permission preset, with:Shell command permission check failed for pattern "!...". Permission for this action has been denied. Reason: Insufficient information about the Bash command to evaluate; action is unverifiable.v4.8.5 tried this approach and crashed/design-engineer:launchfor users in Auto mode. Do NOT use bash injection in command bodies. ${CLAUDE_PLUGIN_ROOT}— officially documented forhooks/hooks.jsoncommandfields ONLY. Does not auto-expand inside slash command markdown bodies. Hooks may use it; commands may not.Skilltool to invoke plugin skills — every plugin skill setsdisable-model-invocation: true, so the Skill tool will reject them withSkill <name> cannot be used with Skill tool due to disable-model-invocation. The only correct way to load a skill isRead ${DESIGN_ENGINEER_PLUGIN_ROOT}/skills/<name>/SKILL.md and follow its instructions inline.
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.
- 2mo ago First seen · 945 lines · 17,303 tokens per session scan D 8d14d9fa9d92
design-engineer-plugin CLAUDE.md is an instructions file published in the GitHub repository volomydyr/design-engineer-plugin (20 stars, last pushed 4mo ago), licensed MIT. It adds 17,303 tokens to every session, about $0.0692 per session on Opus 5.5. A static security scan graded it D with 2 findings (reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
design-engineer-plugin CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| design-engineer-plugin CLAUDE.mdvolomydyr/design-engineer-plugin | Instructions file | 20 | 4mo ago | MIT | D | 17,303 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |