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 noobsplzwin/marketplace-lister-mcp --skill marketplace-listing-copygit clone --depth 1 https://github.com/noobsplzwin/marketplace-lister-mcpWrote 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/noobsplzwin/marketplace-lister-mcp/marketplace-listing-copy)<a href="https://agentmods.dev/skills/noobsplzwin/marketplace-lister-mcp/marketplace-listing-copy"><img src="https://agentmods.dev/badge/skills/noobsplzwin/marketplace-lister-mcp/marketplace-listing-copy/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/noobsplzwin/marketplace-lister-mcp/marketplace-listing-copy"><img src="https://agentmods.dev/badge/skills/noobsplzwin/marketplace-lister-mcp/marketplace-listing-copy.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 | $0.00101 | $0.01610 |
| Opus 5 | $0.00051 | $0.00805 |
| Sonnet 5 | $0.00020 | $0.00322 |
| Haiku 4.5 | $0.00010 | $0.00161 |
Grade A, and why
marketplace-listing-copy 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 12d 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 — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Marketplace listing copy
The one thing to understand first
The grid a buyer scrolls shows photo, title, price. Nothing else. So:
- Title + cover photo + price earn the click.
- Description earns the message. Nobody reads it until they've already clicked.
Write them for those two different jobs. A description packed with specs cannot rescue a title nobody clicks.
Title formula
[Brand] [Product] ([Color] / [Size] / [Measurement]) — [Condition], $[Price]
Rules, in priority order:
- Brand first. It is the most-searched word and it front-loads value.
- Include the one field buyers filter on for that category (see below). Its absence is why a listing gets skipped.
- State condition in the title. Sellers who do this consistently outperform; it is the most commonly omitted field.
- Use the word buyers type, not the word the manufacturer uses. See below.
- Price in the title is optional. Facebook already shows it, so it buys emphasis rather than information.
Most competing listings are titled Chair, Dell monitor, coffee table. A specific title both stands out in the grid and matches far more searches. There is no downside to length.
The searchable-word rule
Title the item the way a buyer would search for it, then add the manufacturer's term if it differs.
Measured on a real local market:
- Searching
dutch ovenreturned 26 listings. Zero said "French Oven." - Searching
french ovendidn't work at all. Facebook doesn't recognize the phrase, degrades to matching "oven", and returns ranges, toaster ovens, and a toy oven.
Staub officially calls its product an "Essential French Oven." A listing titled that way is invisible to everyone searching "dutch oven," which is everyone. Correct title:
Staub 3.6L Cast Iron Dutch Oven (24cm, Cherry Red) — Like New, Made in France
Same trap elsewhere: buyers search "couch" more than "sofa", "TV" more than "television", "pressure cooker" alongside "Instant Pot". When two words compete, put the common one in the title and the precise one in the description. Keep model names that are genuinely searched (Instant Pot, Chuck Taylor).
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.
- 12d ago First seen · 132 lines · 101 tokens per session scan A b3a1e4289a9a
marketplace-listing-copy is a skill published in the GitHub repository noobsplzwin/marketplace-lister-mcp (1 stars, last pushed 19d ago), licensed MIT. It adds 101 tokens to every session and 1,610 once invoked, about $0.0005 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 skills, from other repositories
isuzu-unity-cli
Control Unity Editor from the CLI with the isuzu-unity-cli command. Execute C# code, browse scene hierarchy, inspect/modify GameObjects, capture screenshots, read console logs, check compile status, control play mode, and execute menu items. Use when: user wants to interact with Unity Editor programmatically, run C#…
openclaw
Control a Windows machine remotely via MCP protocol. 40 tools for desktop automation, system management, file operations, and more.
query-superset
Run bounded SELECT queries through an Apache Superset SQL Lab connection on Windows. Use when a user wants an Agent to configure a supported Superset instance, test connectivity, inspect tables or columns, execute SELECT queries, export small results, preserve query evidence, or diagnose authentication, permission…
reality-strategy
Change a blocked real-life or organisational situation by recovering the original result, reconstructing real participants and flows, investigating the sustaining cause, mapping stakeholders and practical power, then altering defaults, burdens, incentives, coalitions, information flow, decision rights or action…
reality-grounding
Actively discover how a situation really works before advice, planning, or system design. Use when the answer may depend on missing facts, actual permissions, stakeholder behaviour, informal practice, available evidence, or feasibility, or when a confused multi-party account may be describing a proposed method instead…
servicenow-safe-deployment
Safely build and deploy ServiceNow artifacts (business rules, scripts, flows, catalog items) across instances using NowAIKit MCP tools, with update sets, dry-run previews, and ATF verification. Use when the user asks to deploy, promote, build, or move changes between dev/test/prod.