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/sigistry/marketplaceWrote 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/commands/sigistry/marketplace/mcp-scaffold)<a href="https://agentmods.dev/commands/sigistry/marketplace/mcp-scaffold"><img src="https://agentmods.dev/badge/commands/sigistry/marketplace/mcp-scaffold/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/commands/sigistry/marketplace/mcp-scaffold"><img src="https://agentmods.dev/badge/commands/sigistry/marketplace/mcp-scaffold.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 | $0.00000 | $0.00955 |
| Opus 5.5 | $0.00000 | $0.00382 |
| Sonnet 5.5 | $0.00000 | $0.00191 |
| Haiku 4.5 | $0.00000 | $0.00096 |
Grade A, and why
mcp-scaffold 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 — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scaffold a new Model Context Protocol server from $ARGUMENTS (server name, language ts|python, transport http|stdio; ask for whatever is missing). Produce a minimal, correct, 2026-07-28-shaped server the user can run and extend, not a kitchen-sink template. Load the mcp-2026-07-28 skill for the transport/lifecycle shape, the mcp-tool-design skill for the example tool, and, for a remote http server, the mcp-authorization and mcp-server-security skills.
Process
Step 1: Confirm the shape
Decide from the arguments and the surrounding repo:
- Language + SDK. TypeScript →
@modelcontextprotocol/sdk(McpServer+StreamableHTTPServerTransport). Python →mcp/fastmcp(FastMCPwith the streamable-HTTP app). Match the repo's existing toolchain if there is one. - Transport.
stdiofor a local, per-user tool launched by the client;http(Streamable HTTP) for a remote, multi-user server. Default tohttpwhen the name implies a hosted service,stdiootherwise. - Stateless by default. Build the HTTP server so each request is self-contained (no reliance on in-process session state that a second replica would not share), per the 2026-07-28 stateless core. See the
mcp-2026-07-28skill.
Step 2: Generate the server
Write a small, working tree. For a TypeScript Streamable-HTTP server that is roughly:
package.json(ESM,@modelcontextprotocol/sdk+zod, astartscript),tsconfig.json,src/server.ts(constructs theMcpServer, registers tools/resources),src/index.ts(Express/Node HTTP wiring theStreamableHTTPServerTransport, stateless mode, health check),.gitignore,.env.example,README.md.- Register one real example tool with a complete zod
inputSchema, correct annotations (readOnlyHintetc.), and a structured error path, followingmcp-tool-design. No placeholderfoo/bar.
For Python, the equivalent pyproject.toml + server.py using FastMCP and its streamable-HTTP app.
Also emit a server.json (registry manifest) stub and, if http, a sample reverse-proxy snippet, so the server is publishable, not just runnable.
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 First seen · 44 lines · 0 tokens per session scan A 35c03a50a406
mcp-scaffold is a command published in the GitHub repository sigistry/marketplace (3 stars, last pushed 4d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 955 tokens. 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 2026-09-03.
Other commands, from other repositories
dotnet-harden
Scan and harden .NET backend code against high-impact anti-patterns such as sync-over-async, lifetime bugs, fat endpoints, and fragile SignalR state.
dotnet-critique
Deep architecture critique of pure .NET backend code. Evaluates AI slop, OOP/SOLID, layer boundaries, DI lifetimes, endpoints, SignalR, data access, concurrency, and distributed-system choices.
py-critique
Deep architecture critique of Python backend code. Evaluates AI slop, SOLID compliance, layer boundaries, anti-patterns, and design quality.
dotnet-structure
Analyze and recommend .NET backend solution and folder structure improvements. Checks projects, layers, oversized files, and boundary clarity.
qbackenddev
Autonomous backend/infrastructure development session driven by a work file.
f5-design
Generate design documents and specifications (D1-D4).