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 skills/nekuda-ai/webmcp-kit/implementnpx skills add nekuda-ai/webmcp-kit --skill implementgit clone --depth 1 https://github.com/nekuda-ai/webmcp-kitWhat 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 | $0.00061 | $0.03408 |
| Opus 5 | $0.00030 | $0.01704 |
| Sonnet 5 | $0.00012 | $0.00682 |
| Haiku 4.5 | $0.00006 | $0.00341 |
Grade A, and why
implement 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 2d 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
WebMCP Kit — implement
Mission
- WebMCP hands a browser agent typed tools (via the
@nekuda/webmcp-sdkSDK) so it answers from the site's own content and acts through the site's own logic. User and agent share the visible page — every tool produces a visible effect. - Tools answer "what will a visitor ask, and ask for, here?" — journeys, never REST-endpoint wrappers.
- In every human-facing run, the plan you present in Phase D is approved by the developer before any file is written. That gate is the product, not a formality.
Hard rules
- Source analysis and code stay local. Never send customer code, routes, or schemas to an external service or to an unauthenticated tool/API. The tool-selection rubric is this skill's text — there is no hosted scanner. The optional post-build Connect flow sends only the CLI's structured project/account request described in
references/connect.md; it is separate from the SDK's documented default-on anonymous observations, which are independent of Connect. - Client-reachable wiring only.
executemay use the app's own client data layer, same-origin routes, or client-safe actions — never server-only imports, secrets, third-party endpoints, or DOM-scraping when a data path exists. - Flag, don't fake. A journey with no safe client path is listed as needs developer wiring — never invented data or a dead call. A guidance-only stub does not "cover" a must-have.
- Irreversible/cost-bearing writes need a boundary. A payment, cancellation, or delete must not complete in one agent call. The generated tool stops at a reversible handoff — creates the repo-native pending state (checkout session, pending order, prepared cancellation) and hands the final step to the app's own payment/confirm UI, or uses a prepare→confirm two-call shape. A consequence sentence in the description is not a boundary.
- Authorization is the server's job. Registration gating on auth/role is UX, not security. Only wrap a privileged mutation whose route independently enforces authn/authz server-side; an endpoint that trusts a hidden client button is needs developer wiring, not a tool.
- No changes or Connect before approval. No branch, file, dependency, or Connect attempt until the developer approves the plan, except the tool-generation workflow in explicit
--non-interactivemode; that exception permits the branch, files, and dependencies needed for generation but never Connect, which non-interactive mode skips. This skill reads with Read/Grep/Glob during A–C — plus one read-only exception: the Phase-F baseline capture (the repo's own typecheck/build,references/verify.md) may run as a background task before approval, because it writes nothing to the tree; after approval it requests the write, package-manager, and browser permissions it needs through the normal permission prompts — no mutation is pre-authorized. Interactive Connect is a separate, optional post-build decision (references/connect.md). - SDK only. Generated code imports
defineTool/registerToolsfrom the SDK under its resolved name (the installed package's declared name —references/sdk.md); never the rawmodelContextsurface, never a bundled polyfill/shadow. The SDK pins the spec, resolves whichever surface the browser exposes, and no-ops when unsupported.
What ships with it
13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- interactive/ack-event.ts 14 KB runs code
- interactive/explorer.html 167 KB
- interactive/protocol.ts 12 KB runs code
- interactive/server.ts 24 KB runs code
- references/codegen.md 9.3 KB
- references/connect.md 7.7 KB
- references/interactive.md 14 KB
- references/journeys.md 8.1 KB
- references/plan-template.md 2.6 KB
- references/sdk.md 3.8 KB
- references/verify.md 6.2 KB
- references/wiring.md 3.2 KB
- scripts/star-repo.sh 1.4 KB runs code
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.
- 2d ago First seen · 63 lines · 61 tokens per session scan A 218a8a9a023b
implement is a skill published in the GitHub repository nekuda-ai/webmcp-kit (24 stars, last pushed 7d ago), licensed MIT. It adds 61 tokens to every session and 3,408 once invoked, about $0.0003 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 skills, from other repositories
webmcpify
WebMCP agent skill for curated core coverage or route-by-route parity — inventory an existing web app, integrate approved tools, then verify and heal them in a real browser. Use for "webmcpify", "add WebMCP", or "expose app actions to AI agents".
clean-user-facing-text
Clean and finalize authorized natural-language text intended for readers by auditing suspicious invisible Unicode and rewriting prose while preserving facts, meaning, and the writer's voice. Use when the user asks to clean, humanize, polish, or finalize articles, manuscripts, reports, documentation, emails, product…
proxies
Oxylabs proxy networks: Residential, Mobile, shared Datacenter/ISP, and Dedicated Datacenter/ISP proxies with geo-targeting, IP rotation, session persistence, and port-based sticky IPs. Use when routing traffic through proxies, building scrapers with proxy auth, rotating or sticky sessions, whitelisting IPs, or…
karpathy-llm-wiki
Use when building or maintaining a personal LLM-powered knowledge base. Triggers: ingesting sources into a wiki, querying wiki knowledge, linting wiki quality, 'add to wiki', 'what do I know about', or any mention of 'LLM wiki' or 'Karpathy wiki'.
agentsop-map-reduce-fanout
Decision protocol for the map-reduce / dynamic fan-out pattern in LM pipelines — "given list L, run f(item) for each item in parallel, then combine". Activates when the coder agent is about to process N items with N LM calls (per-doc summarize, per-query retrieve, per-candidate rank, parallel tool fan-out). Encodes…
agentsop-prompt-history-inspect
Tool skill — the first move in any LM-debugging session: dump the actual rendered prompt the framework sent to the model, before changing anything else. Activate when an LM call produced an unexpected output (wrong answer, schema violation, refusal, truncation, cost spike, latency spike, infinite loop, "model got…