oahelper-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| OAHELPER_HOME | No | Data folder for .env, cache/ and library/. Defaults to ~/.oahelper, or the clone folder when run from a clone. Environment only. | |
| OAHELPER_BEARER | No | The authorization header value, if the site sends one. | |
| OAHELPER_COOKIE | Yes | Browser cookie header from a logged-in OAHelper session. Paste the whole cookie value from a request whose URL contains /api/proxy/. Required to access your Pro account. | |
| OAHELPER_DAILY_CAP | No | Live requests per day before the server refuses. Default is 300. | |
| OAHELPER_USER_AGENT | No | Your browser's user agent, so requests match your session (default is a generic user agent). |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| oahelper_searchA | Find a company's OA question sets on OAHelper by name (e.g. "Amazon", "Kickdrum"). Lists each set with its company_ref, date and question counts. To download a company's sets into the offline library, use oahelper_sync. |
| oahelper_syncA | Download one company's OAHelper content into the offline library, newest sets first: set lists, every question statement, MCQ answers, and interview experiences. Makes at most max_items live requests per call, then says how much is left, so call it again to continue. Already-saved items cost nothing. Official solutions are NOT fetched here because each one uses the account's daily solution quota; use oahelper_solution. Read the result from library//index.md. |
| oahelper_companyA | List one OAHelper set page by slug or URL, e.g. "kickdrum-mcq24sep" or https://www.oahelper.in/company-questions/kickdrum-mcq24sep. Returns question refs for oahelper_question. For a company by name, use oahelper_search or oahelper_sync. |
| oahelper_questionA | Read one OAHelper question: full statement, MCQ answers with explanations (for MCQ sets), any saved official solution, and optionally community solutions and images. Saves it to the offline library. ref_or_url: a question ref like "MjM3OHww" or a https://www.oahelper.in/problems/... URL. |
| oahelper_solutionA | Fetch OAHelper's official solution for one coding question (language: cpp, python, java). Each new fetch uses one of the account's limited daily solution requests, so only call this when the user asks for the official solution. Saved solutions are free. |
| oahelper_interviewsA | Interview experiences for a company from OAHelper (12 per page, most relevant first). Saves each full write-up to library//interviews/ and returns a short list; read the saved files for the full text. |
| oahelper_offlineA | What is saved offline. No network call. With no company, lists every company in the library; with a company, returns its index (sets, interviews, counts). |
| oahelper_statusA | Session and usage status. online=true also checks Pro status with the site (one request); by default no network call is made. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
Most tools target distinct actions (search companies, sync a company, read a question, fetch a solution, list interviews, inspect offline library, check status). The only mild overlap is oahelper_search vs oahelper_company, both of which locate sets, but their descriptions explicitly disambiguate name-based vs slug/URL-based lookup. The bulk-vs-single distinction between oahelper_sync and oahelper_question/solution is also spelled out.
All tools use a uniform oahelper_ snake_case prefix, which makes the set predictable and easy to scan. The only slight deviation is mixing action verbs (search, sync, offline) with domain nouns (company, question, solution, interviews, status), but formatting itself is consistent.
Eight tools is well-scoped for an OA interview-prep domain. Each tool earns its place: discovery, retrieval, bulk sync, offline inspection, and status each cover a distinct need without redundancy.
The surface covers the full lifecycle: find companies, inspect sets, sync content offline, read questions, fetch solutions, view interviews, and check offline/session state. Minor gaps remain, such as no direct standalone question-search or explicit login/auth tool, but status and search partially cover these.