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/ashishpatill/tell-proofWrote 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/rules/ashishpatill/tell-proof/tell-site-build-autoload)<a href="https://agentmods.dev/rules/ashishpatill/tell-proof/tell-site-build-autoload"><img src="https://agentmods.dev/badge/rules/ashishpatill/tell-proof/tell-site-build-autoload/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/rules/ashishpatill/tell-proof/tell-site-build-autoload"><img src="https://agentmods.dev/badge/rules/ashishpatill/tell-proof/tell-site-build-autoload.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 | $0.00468 | $0.00468 |
| Opus 5 | $0.00234 | $0.00234 |
| Sonnet 5 | $0.00094 | $0.00094 |
| Haiku 4.5 | $0.00047 | $0.00047 |
Grade A, and why
tell-site-build-autoload 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 11d 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
Site build auto-load
When the user asks to build, redesign, or prepare any website / landing / matchday / specimen site, execute the full skill graph without waiting to be asked to wire or trigger individual skills.
Auto-run order (mandatory)
- Research —
website-domain-researchend-to-end
(load-prior-domain→requirement-gap-diff→ walkthrough if gaps →ia-shell-synthesis→category-gap-audit→variant-lens→emit-training-episode). - Sport — if cricket / football / hockey / tennis / live scores / matchday: also
sport-matchday-web+sport-site-research+sport-vernacular-craft. - Craft — execute every node returned by engine
routeSkills/ niche craftNodes (including always-onresponsive-performance, motion, nav/footer, etc.). Do not skip to pixels after research alone. - Media performance (automatic) — after any photography lands under
apps/web/public/**:- Run
pnpm media:site(WebP + display-size budgets) - Prefer
--pruneso hot-path jpg/png do not ship - Render with
SiteImg(lazy below-fold, LCP priority on one hero, width/height set)
- Run
- Agency path —
agency:run/agency-quality-sitewhen shipping a phased site; learn is automatic. Media optimize is part of ship readiness.
Do not
- Blank-slate when a
DomainResearchPackexists - Ship multi-MB JPEG/PNG heroes
- Wait for the user to say “run research”, “convert to webp”, or “wire responsive-performance”
- Treat
routeSkillslisting as done — execute the nodes
Engine: @tell/design-skills routeSkills / routeDomainResearchSkills / designFromFeatures.
Skills live in .cursor/skills/ and agent-skills/web-design/.
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.
- 11d ago First seen · 39 lines · 468 tokens per session scan A 1c9f7205420d
tell-site-build-autoload is a cursor rule published in the GitHub repository ashishpatill/tell-proof (1 stars, last pushed 16d ago), licensed MIT. It adds 468 tokens to every session, about $0.0023 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 2026-08-31.
Other cursor rules, from other repositories
graphql-typescript-integration
A TypeScript rule for GraphQL operations that uses the API schema and queries to generate types. It helps type-check data and variables from the GraphQL server through to React components.
observer-hoc-or-useobserver-hook
Guidance for making React components respond to changes in MobX observable data by wrapping them with an observer or using the useObserver Hook.
graphql-apollo-client-usage
A usage guide for fetching, changing, and subscribing to GraphQL data with Apollo Client in React. It covers loading and error states, cache updates, refetching, and reusable GraphQL fragments.
chakra-ui---component-composition
A React component-structure rule for composing Chakra UI components into reusable interfaces. It favors small custom components built from layout and interface pieces such as Box, Flex, and Stack.
chakra-ui---theme-directory-rules
Rules for organizing a Chakra UI theme into an entry file, foundational settings, and component-specific styles. The theme controls shared colors, fonts, spacing, breakpoints, and component variants.
chakra-ui-best-practices
A set of rules for building React interfaces with Chakra UI, a component library for styled and accessible user interfaces. It covers themes, responsive layouts, accessibility, performance, and reusable components.