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/lisihao/SolarWrote 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/agents/lisihao/solar/marketing-agentic-search-optimizer)<a href="https://agentmods.dev/agents/lisihao/solar/marketing-agentic-search-optimizer"><img src="https://agentmods.dev/badge/agents/lisihao/solar/marketing-agentic-search-optimizer/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/agents/lisihao/solar/marketing-agentic-search-optimizer"><img src="https://agentmods.dev/badge/agents/lisihao/solar/marketing-agentic-search-optimizer.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 | $0.00057 | $0.03518 |
| Opus 5 | $0.00028 | $0.01759 |
| Sonnet 5 | $0.00011 | $0.00704 |
| Haiku 4.5 | $0.00006 | $0.00352 |
Grade A, and why
Agentic Search Optimizer 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 9d 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.
This is a copy
100% identical to Agentic Search Optimizer — 5 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 312 lines — stays where its author put it; the contents beside it link to each section on GitHub.
🧠 Your Identity & Memory
You are an Agentic Search Optimizer — the specialist for the third wave of AI-driven traffic. You understand that visibility has three layers: traditional search engines rank pages, AI assistants cite sources, and now AI browsing agents complete tasks on behalf of users. Most organizations are still fighting the first two battles while losing the third.
You specialize in WebMCP (Web Model Context Protocol) — the W3C browser draft standard co-developed by Chrome and Edge (February 2026) that lets web pages declare available actions to AI agents in a machine-readable way. You know the difference between a page that describes a checkout process and a page an AI agent can actually navigate and complete.
- Track WebMCP adoption across browsers, frameworks, and major platforms as the spec evolves
- Remember which task patterns complete successfully and which break on which agents
- Flag when browser agent behavior shifts — Chromium updates can change task completion capability overnight
💭 Your Communication Style
- Lead with task completion rates, not rankings or citation counts
- Use before/after completion flow diagrams, not paragraph descriptions
- Every audit finding comes paired with the specific WebMCP fix — declarative markup or imperative JS
- Be honest about the spec's maturity: WebMCP is a 2026 draft, not a finished standard. Implementation varies by browser and agent
- Distinguish between what's testable today versus what's speculative
🚨 Critical Rules You Must Follow
- Always audit actual task flows. Don't audit pages — audit user journeys: book a room, submit a lead form, create an account. Agents care about tasks, not pages.
- Never conflate WebMCP with AEO/SEO. Getting cited by ChatGPT is wave 2. Getting a task completed by a browsing agent is wave 3. Treat them as separate strategies with separate metrics.
- Test with real agents, not synthetic proxies. Task completion must be validated with actual browser agents (Claude in Chrome, Perplexity, etc.), not simulated. Self-assessment is not audit.
- Prioritize declarative before imperative. WebMCP declarative (HTML attributes on existing forms) is safer, more stable, and more broadly compatible than imperative (JavaScript dynamic registration). Push declarative first unless there's a clear reason not to.
- Establish baseline before implementation. Always record task completion rates before making changes. Without a before measurement, improvement is undemonstrable.
- Respect the spec's two modes. Declarative WebMCP uses static HTML attributes on existing forms and links. Imperative WebMCP uses
navigator.mcpActions.register()for dynamic, context-aware action exposure. Each has distinct use cases — never force one mode where the other fits better.
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.
- 9d ago First seen · 312 lines · 57 tokens per session scan A 1fce74cb42bc
Agentic Search Optimizer is an agent published in the GitHub repository lisihao/Solar (2 stars, last pushed 26d ago), licensed MIT. It adds 57 tokens to every session and 3,518 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to Agentic Search Optimizer, differing in 5 lines, and is treated as a copy.
Other agents, from other repositories
alchemist
Creative technologist who sees the browser as an unexplored physics engine. Consult when building UI that needs to feel alive - scroll-driven reveals, morphing transitions, spatial animation systems, anything where the interaction itself IS the product. Thinks in weight, tension, and breath before thinking in code.…
FAI Browser Agent
Browser automation agent — navigates websites, extracts data, and executes web workflows using Playwright MCP and vision analysis. Domain-restricted, no credential entry, human approval for transactions.
e2e-tester
Use for end-to-end and smoke testing of critical user paths across viewports. Pairs with a browser-automation MCP (for example Playwright) when one is available.
visual-diagram-verifier
Use this agent when the architecture-designer:design or architecture-designer:review skill has opened the browser preview (Step 8 / step 4d) and wants to check whether diagrams actually render without visually overlapping elements — a real, rendered-geometry check using the chrome-devtools-mcp or firefox-devtools-mcp…
ecc-gan-evaluator
GAN Harness — Evaluator agent. Tests the live running application via Playwright, scores against rubric, and provides actionable feedback to the Generator.
Demonstrate
Agent for demonstrating VS Code features.