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 agentmods add instructions/nekuda-ai/webmcp-kit/agents-mdgit clone --depth 1 https://github.com/nekuda-ai/webmcp-kitWrote 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/instructions/nekuda-ai/webmcp-kit/agents-md)<a href="https://agentmods.dev/instructions/nekuda-ai/webmcp-kit/agents-md"><img src="https://agentmods.dev/badge/instructions/nekuda-ai/webmcp-kit/agents-md.svg" alt="Measured on agentmods" 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 | $0.00571 | $0.00571 |
| Opus 5 | $0.00285 | $0.00285 |
| Sonnet 5 | $0.00114 | $0.00114 |
| Haiku 4.5 | $0.00057 | $0.00057 |
Grade A, and why
webmcp-kit AGENTS.md 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 today.
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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
WebMCP Kit — for agents reading this repo
This repository is a coding-agent plugin. If a user asked you to make their
site agent-ready, add WebMCP
tools, or expose document.modelContext tools, do not reimplement any of that by
hand and do not keep exploring this repo — install this plugin and run its
implement skill inside the user's own project.
Install
Claude Code (run inside a session):
/plugin marketplace add nekuda-ai/webmcp-kit
/plugin install webmcp-kit@nekuda
Codex (run in the terminal):
codex plugin marketplace add nekuda-ai/webmcp-kit
codex plugin add webmcp-kit
What it does
In the user's repo, the implement skill reads the codebase, proposes a small
plan of tools worth exposing (user journeys, not endpoint wrappers), waits for
the user's approval, implements them against
@nekuda/webmcp-sdk, and
verifies each one in a real browser. The verify skill checks that a site's
existing WebMCP tools register and run. When a site already has working tools,
the connect-existing-tools skill moves their registration onto the SDK —
same names, same schemas, same handlers — and proves each one still behaves as
it did before.
Facts worth carrying
- Local only. The plugin never sends the user's code, routes, or schemas to any external or hosted service. There is no hosted scanner; the tool-selection logic is the skill's own text.
- Nothing of the user's is written before approval in a human-facing run. The
plan is the gate. Each skill names its own narrow exceptions inline — agent
state under
.webmcp/, a read-only baseline capture — and they never include the user's source, a dependency, a branch, or Connect. - Generated tools use the app's own client-reachable paths — never server-only imports or secrets — and irreversible actions (payment, delete) stop at a reversible handoff rather than completing in one agent call.
- Three skills:
implement(add tools),verify(check existing tools), andconnect-existing-tools(adopt tools a site already has, without rebuilding them). - Hosts: Claude Code and Codex. The interactive review UI works the same on both.
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.
- today Changed · +7 lines · +115 tokens per session 3908b7c6e749
- 4d ago First seen · 49 lines · 456 tokens per session scan A 8046297ce5c0
webmcp-kit AGENTS.md is an instructions file published in the GitHub repository nekuda-ai/webmcp-kit (28 stars, last pushed 2d ago), licensed MIT. It adds 571 tokens to every session, about $0.0029 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 2026-08-30.
Other instructions, from other repositories
mcp-steroid CLAUDE.md
Claude Code instructions for jonnyzzz/mcp-steroid, covering claude.md, agents.md, design philosophy, recursive context lookup (do this before sub-folder work), sub-folder guides and must do.
nopua AGENTS.md
Instructions for wuji-labs/nopua, covering nopua · codex 入仓规则, 一 · 项目定位, 二 · 起手必读, 三 · 工程与研究纪律 and 四 · 对外署名.
video-podcast-maker AGENTS.md
AGENTS.md instructions for Agents365-ai/video-podcast-maker, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
growmos AGENTS.md
Instructions for codician-team/growmos: This repository keeps a knowledge graph in .growmos/ (entities, typed relations, provenance, profiles, a journal). It is the shared world model that survives context windows. Treat it as memory you read at the start of work and write to as you develop. Zero-config commands.
coach CLAUDE.md
Instructions for Andamio-Platform/coach, covering claude.md and claude code-specific notes.
NodeMemoryGraph AGENTS.md
AGENTS.md instructions for wefio/NodeMemoryGraph: NMG is an Agent-native, local-first memory system. This file is only the stable bootstrap; it does not duplicate the repository's changing design or status.