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 skills add Noibu/ai-plugin --skill tech-diagnosisgit clone --depth 1 https://github.com/Noibu/ai-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/skills/noibu/ai-plugin/tech-diagnosis)<a href="https://agentmods.dev/skills/noibu/ai-plugin/tech-diagnosis"><img src="https://agentmods.dev/badge/skills/noibu/ai-plugin/tech-diagnosis/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/skills/noibu/ai-plugin/tech-diagnosis"><img src="https://agentmods.dev/badge/skills/noibu/ai-plugin/tech-diagnosis.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 | $0.00075 | $0.07378 |
| Opus 5.5 | $0.00030 | $0.02951 |
| Sonnet 5.5 | $0.00015 | $0.01476 |
| Haiku 4.5 | $0.00007 | $0.00738 |
Grade A, and why
tech-diagnosis 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 — 424 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tech Diagnosis
- Proactive — no specific issue. Scans for highest-value errors and CWV problems, surfaces a prioritized list.
- Reactive — a specific signal is named. Produces a self-contained finding (Summary → Cause → Fix).
Call metadata: on every noibu_* call this skill drives, pass skill: "tech-diagnosis" and a short skillStep (see the querying-noibu-data skill).
Setup (always runs first)
- Noibu MCP — confirm connected. If not, stop.
- Domain — UUID → use it. Name only → resolve. Nothing → list and ask. Capture platform (Shopify, Magento, etc.).
- Mode + window — detect mode (see below). Default window: 30 days.
- TodoList — create with mode-appropriate tasks.
There is no config file. Connector status is detected at runtime by checking tool availability. Preferences are asked per session.
Mode detection
- Proactive — no specific issue given. Domain name alone is proactive. Examples: "run tech diagnosis", "what's broken", "scan for issues".
- Share with vendor — message starts with "Share with vendor:". Skip setup and diagnosis → go to Step P4 Share with vendor.
- Ignore issue — message starts with "Ignore issue #". Skip setup and diagnosis → go to Ignore issue flow.
- Reactive — specific issue named alongside domain: error name, metric + page, Noibu issue ID, page URL, or structured handoff. Also reactive for connector reconfiguration requests.
- Ambiguous — ask: "Are you looking for a scan of what's broken across the store, or do you have a specific issue in mind?"
Rules
- No client-side data joins. Each Noibu source on its own terms.
- Don't overstate impact. "Present in X sessions", not "blocking X users". No revenue projections or revLost figures.
- Plain language. Translate Noibu labels; keep identifiers in code style.
- Suppress citation sections.
- First visible output is the result, never narration. No preambles, filler, or "let me pull X".
- Never narrate internals. No tool names, query parameters, API state values, filter logic, or fallback decisions. Retries happen silently.
- Pause only at designated points. Reactive: the connector check before diagnosis (Stage 1, plus Stage 2 only when GitHub was declined), then once after the closing offer widget. Proactive: after findings widget. Open-ended questions only; never AskUserQuestion modals.
- After every
show_widgetcall, output one line only. Proactive: "Select an option on a card above to get started." Reactive (closing offer): "Select an option above, or choose another issue to fix." No summary, no recap, nothing else. - Page criticality is internal only. Checkout > Cart > PDP > Collection > Home > Other. Never a labeled field.
- Name third-party services specifically. Stack trace says Klaviyo, GTM, Shop Pay → use that name.
- Never render a widget from memory. If resuming from a compacted conversation, re-read this SKILL.md before any output.
What ships with it
11 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/errors.md 4.1 KB
- references/first-run-setup.md 6.7 KB
- references/performance.md 4.1 KB
- references/platforms/bigcommerce.md 1.4 KB
- references/platforms/generic.md 2.1 KB
- references/platforms/magento.md 1.5 KB
- references/platforms/shopify.md 2.7 KB
- references/platforms/woocommerce.md 1.7 KB
- references/schedule-widget.md 15 KB
- references/share.md 2.2 KB
- references/ticket.md 2.8 KB
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 Changed · +2 lines 5b252914cdbc
- 2mo ago First seen · 422 lines · 75 tokens per session scan A 2782c9f1a0d4
tech-diagnosis is a skill published in the GitHub repository Noibu/ai-plugin (5 stars, last pushed 13d ago), licensed MIT. It adds 75 tokens to every session and 7,378 once invoked, about $0.0003 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 2026-08-30.
tech-diagnosis compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| tech-diagnosisNoibu/ai-plugin | Skill | 5 | 13d ago | MIT | A | 75 | Claude Code |
| revenuecat-troubleshootRevenueCat/ai-toolkit | Skill | 79 | 2d ago | MIT | A | 125 | Claude Code |
| b2c-logsSalesforceCommerceCloud/b2c-developer-tooling | Skill | 58 | today | Apache-2.0 | A | 100 | Claude Code |
| vendure-troubleshootingvendurehq/vendure | Skill | 8.6k | yesterday | none | A | 89 | Claude Code |
| doctorHainrixz/agente-pagokit | Skill | 70 | 17d ago | MIT | A | 105 | Claude Code |
| b2c-production-triageSalesforceCommerceCloud/b2c-developer-tooling | Skill | 58 | today | Apache-2.0 | A | 58 | Claude Code |
| b2c-checkout-triageSalesforceCommerceCloud/b2c-developer-tooling | Skill | 58 | today | Apache-2.0 | A | 49 | Claude Code |