openings-mcp: Skill for Claude Code

.agents/skills/integrate-new-provider/SKILL.md

integrate-new-provider is a skill for Claude Code, Codex from amikai/openings-mcp. It costs 104 tokens per session (3,059 once invoked), scanned A, original, MIT.

A workflow for connecting a new job-listings website or hiring platform to openings-mcp, a server that provides job-opening data to other software.

In plain words
What is it for?
Use it to inspect a provider, capture test data, build its client and package, add a debugging command, and expose it through the MCP server.
Why use it?
It removes the guesswork involved in finding a provider’s real data source and wiring it into the server with tests.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions CLAUDE.md; mentions Claude Code; installed under .agents/ (shared by several agents).

This is amikai/openings-mcp's own configuration. It tells Claude Code and Codex how to work on openings-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything openings-mcp configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is sleep 15) | go run ./cmd/openings-mcp.

Reuse

Borrowing it

Nothing to install: this file belongs to amikai/openings-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/amikai/openings-mcp/main/.agents/skills/integrate-new-provider/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/amikai/openings-mcp

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for integrate-new-provider

README.md
[![agentmods](https://agentmods.dev/badge/skills/amikai/openings-mcp/integrate-new-provider/github.svg)](https://agentmods.dev/skills/amikai/openings-mcp/integrate-new-provider)
Your own site
<a href="https://agentmods.dev/skills/amikai/openings-mcp/integrate-new-provider"><img src="https://agentmods.dev/badge/skills/amikai/openings-mcp/integrate-new-provider/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.

agentmods 80×15 button for integrate-new-provider

Your own site · 80×15
<a href="https://agentmods.dev/skills/amikai/openings-mcp/integrate-new-provider"><img src="https://agentmods.dev/badge/skills/amikai/openings-mcp/integrate-new-provider.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 104 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,059 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00104 $0.03059
Opus 5 $0.00052 $0.01529
Sonnet 5 $0.00021 $0.00612
Haiku 4.5 $0.00010 $0.00306

Measured 11d ago against content hash e6fddc10b4bc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

integrate-new-provider 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 11d 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.

.agents/skills/integrate-new-provider/SKILL.md · 223 lines

How it starts

The opening of the file, as written. The whole thing — 223 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Integrate a New Provider

Overview

Every provider follows the same pipeline: recon (find and rank the site's real data surface) → capture fixtures → build a client that matches that surface → provider package with fixture-replaying tests → debug CLI → MCP surface (ATS adapter or dedicated tools, wired into the server).

For JSON REST, the client path is: minimal OpenAPI spec → ogen-generated client. For GraphQL, HTML, SSR-embedded state, RSS/Atom, or JSON Feed, write a hand-rolled client instead — do not force ogen. Work the stages in order; each builds on the previous one's verified output. The integration is done when the provider is reachable through the MCP server, not when the debug CLI works — if a session stops before the surface stage, hand off the remaining stages explicitly.

SmartRecruiters (internal/provider/smartrecruiters, cmd/smartrecruiters) is the most recent ogen/JSON worked example. Spec-less hand-written clients in-tree: LinkedIn, jobindex, join, iCIMS, SuccessFactors, UltiPro (HTML / GraphQL / __NEXT_DATA__).

Pick the MCP Surface

  • Multi-company ATS (one API, many tenants/boards): implement internal/ats.Adapter so companies join the unified search_jobs_by_company tools. Examples: workday, greenhouse, lever, ashby.
  • Single site or job board: dedicated <name>_search_jobs / <name>_get_job_detail MCP tools in internal/openingsmcp/<name>.go. Examples: job104, cake, google, tsmc, linkedin.

RSS/Atom and most niche boards land on dedicated tools unless each tenant has a stable, roster-able feed URL and multi-company routing is worth it.

Pipeline

  1. Spec hunt — WebSearch for an official OpenAPI/Swagger spec or developer API docs before reverse-engineering anything (e.g. " API openapi spec", " developer docs"). An official spec beats a hand-derived one: fewer wrong guesses about types, nullability, and pagination. If one exists, trim it down to the endpoints you need rather than writing from scratch. Official or not, the spec still gets verified against captured traffic in the next step — vendor specs drift from what the public endpoints actually return. No official docs does not mean no integration; it means recon (step 2) owns the surface choice.

Read the full file on GitHub · 223 lines

Changes

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.

  1. 11d ago First seen · 223 lines · 104 tokens per session scan A e6fddc10b4bc

Subscribe to this mod's changes

integrate-new-provider is a skill published in the GitHub repository amikai/openings-mcp (92 stars, last pushed 17d ago), licensed MIT. It adds 104 tokens to every session and 3,059 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 2026-08-30.

Related

Other skills, from other repositories

add-sdk-method

Use when adding a new method, function, or API endpoint to the SDK. Ensures consistent implementation across all SDKs (Python, TypeScript, Java, Rust CLI) with proper types, tests, and naming conventions.

open-metadata/ai-sdk · 49 tokens

api-context

Canonical reference for the unified Context object passed to every tool and resource handler in @cyanheads/mcp-ts-core. Covers the full interface, its RequestContext base, all sub-APIs (ctx.log, ctx.state, ctx.requestInput, ctx.inputs, ctx.enrich, ctx.content), and when to use each.

cyanheads/pubchem-mcp-server · 79 tokens

api-errors

McpError constructor, JsonRpcErrorCode reference, and error handling patterns for @cyanheads/mcp-ts-core. Use when looking up error codes, understanding where errors should be thrown vs. caught, or using ErrorHandler.tryCatch in services.

cyanheads/pubchem-mcp-server · 54 tokens

add-service

Scaffold a new service integration. Use when the user asks to add a service, integrate an external API, or create a reusable domain module with its own initialization and state.

cyanheads/pubchem-mcp-server · 38 tokens

api-workers

Cloudflare Workers deployment using createWorkerHandler from @cyanheads/mcp-ts-core/worker. Covers the full handler signature, binding types, CloudflareBindings extensibility, runtime compatibility guards, and wrangler.toml requirements.

cyanheads/pubchem-mcp-server · 51 tokens

add-app-tool

Scaffold an MCP App tool + UI resource pair. Use when the user asks to add a tool with interactive UI, create an MCP App, or build a visual/interactive tool.

cyanheads/pubchem-mcp-server · 41 tokens