FoundRole — AI Job Search & Application Tracker MCP for Claude & ChatGPT

Lists the authenticated user's job alerts across all subscription sources (regular, company page, MCP). **Input:** - `status`: Filter by status — one of pending, active, unsubscribed (optional, default: all statuses) - `limit`: Number of results to return (default 20, max 50) - `offset`: Number of results to skip (default 0) **Output:** Returns the user's job alerts with pagination info and a summary of the underlying job search (query, location, company where available). Each response includes a system_instruction describing how to present the results for the current client.

Subscribes the authenticated user to job alerts for a specific saved job search. **Input:** - `job_search_id`: The job search identifier to subscribe to (required). Accepts either the job search UUID or the composite job ID returned by `jobs_search` / `jobs_details` (format: "seo_id--job_search_id"). - `frequency`: Alert frequency — one of daily, weekly, monthly (optional, defaults to "weekly") **Output:** Returns the created or updated job alert with id, status, and frequency. Idempotent: calling this tool for an already-subscribed search updates the existing alert without creating a duplicate.

Unsubscribes the authenticated user from job alerts for a specific job search. **Input:** - `job_search_id`: The job search identifier to unsubscribe from (required). Accepts either the job search UUID or the composite job ID returned by `jobs_search` / `jobs_details` (format: "seo_id--job_search_id"). **Output:** Confirms the alert has been unsubscribed. Idempotent: returns success even when the user was not subscribed or is already unsubscribed.

Unsubscribes the authenticated user from ALL of their job alerts at once, across every subscription source (regular, company page, MCP). **Input:** - `confirm`: Must be `true` to execute. The call is rejected when omitted or not true — this guards against an unintended bulk unsubscribe. **Output:** Confirms how many alerts were unsubscribed. Idempotent: returns success even when the user has no active alerts.

Analyzes one job found outside FoundRole using the authenticated user's FoundRole profile and the same signals used for FoundRole jobs: resume match, missing skills, H-1B sponsorship history, E-Verify, ghost-job risk, posted compensation, and market salary estimates. Use tracker_add_external only when the user asks to save without analysis. The input represents the direct posting URL and all job content already available in the conversation. The five text identity fields are required; every structured fact field is optional, with a fact the source does not state simply omitted (or null). The optional `client_extraction` object carries evidence-backed skills, technology, benefits, bonuses, seniority, industry, management, clearance, visa, and remote-scope labels when source excerpts for them exist. FoundRole validates the evidence, stores the client extraction separately, derives missing deterministic facts, and reports which values were provided, derived, accepted, rejected, or remain unknown. The output includes `comparisonRef`; retain it exactly for a later jobs_compare call. The analysis is a decision aid, not a guarantee about sponsorship, legitimacy, compensation, or hiring outcome.

Compares 2 to 4 jobs side by side using the same FoundRole analysis fields: resume match, missing skills, H-1B and E-Verify signals, ghost-job risk, posted pay, and market salary estimates. `comparison_refs` accepts exact FoundRole job IDs returned by jobs_search and exact external `comparisonRef` URLs returned by jobs_analyze_external. Analyze each outside job first; a bare URL that has not been analyzed cannot be compared because FoundRole does not have its posting facts. Preserve every reference exactly, keep the user's requested order, and do not send duplicates.

Fetches full details for one job by the `id` returned from jobs_search — the deeper view behind a search result. **Input:** - `job_id`: The exact ID string from the `id` field of a `jobs_search` result. **Output:** Complete job details: description, skills, benefits, requirements, salary benchmark, resume match, H-1B and E-Verify signals, job-trust analysis, and application link. Personalized and extended insight fields follow the authenticated user's current entitlements.

Returns the authenticated user's personalized job recommendations built from their resume, skills, target roles, and preferred location. Results are ranked by fit, may include related roles, and carry the same salary, match, H-1B, and job-trust insight payload used by job search. A processing status means the personalized feed is still being prepared; a later call returns the completed feed. Page numbers fetch additional recommendations from the same feed.

Searches a database for real-time job listings matching the user's criteria. The query is the full job title or role: "Ruby Developer" or "Ruby on Rails Engineer" rather than a bare keyword like "Ruby", which is too broad and matches unrelated fields. Results may be filtered by location, company, and how recently a job was posted. Each result carries an `id`; jobs_details takes that `id` and returns the job's full description, requirements, and benefits. The response also carries a `nextCursor` for the next page of results; a follow-up page is fetched by passing only that cursor, with no other search parameters. Authenticated results include resume match data when a profile is available. Job details include FoundRole salary benchmarks, H-1B sponsorship signals, E-Verify status, and job-trust analysis; list-level employer signals follow the user's current entitlements. Advanced constraints in the user's request — remote-only work, H1B sponsorship, a minimum salary, hiding risky postings, a minimum match score — are the search parameters remote, h1b_sponsors_only, salary_floor, hide_low_quality, and min_match. The search enforces only constraints passed as parameters; a constraint left out of the call is not applied to the result set. FoundRole Pro accounts receive the screened list; other accounts receive the full list and the response reports that the advanced filters were not applied. Each response includes a system_instruction describing how to present the results for the current client.

Searches FoundRole's published content by semantic similarity and returns the most relevant sources for a job-search question: career-guidance blog articles plus FoundRole site pages that describe the product's features (job tracker, Pro plan and pricing, H1B salary data, AI job search) and industry/sector career landings. Each article carries a title, url, summary, a content excerpt, publication date, and tags; each page carries a title, url, description, and its FAQ entries — enough material to answer the question and link the source. Three optional facets add further result groups: company returns FoundRole's employer profile pages matching that company name; job_title and location return the live job-listing landing pages for that role and place, with open-job counts. The facets describe what the user is asking about — a company mentioned only in passing does not need the company facet. Returns empty groups when nothing is relevant rather than padding with off-topic content. Results are the closest matches to the given question, not an index of the site's full coverage; questions about overall topic coverage are answered by knowledge_topics, which lists the blog's categories and tags with article counts. It does not search job listings; jobs_search covers live roles. Each response includes a system_instruction describing how to present the sources.

Lists what FoundRole's published career-guidance blog covers: every category and the most-used tags, each with its published-article count and url, plus the total number of published articles. This is the factual source for questions about the blog's topics or overall coverage. It takes no parameters and reflects the live published corpus. It does not retrieve articles for a specific question; knowledge_search does that.

Deletes a reminder from a tracked job. **Input:** - `tracked_job_id`: The tracked job ID — `trackedJobs[].id` from tracker_list output, distinct from `trackable.id` and `job.id` (required) **Output:** Returns the updated tracked job with reminderAt cleared.

Lists tracked jobs that have reminders set, ordered by reminder time (soonest first). **Input:** - `limit`: Number of results to return (default 20, max 50) **Output:** Returns a list of tracked jobs with active reminders.

Sets a reminder for a tracked job. Sends a confirmation email with .ics calendar attachment. **Input:** - `tracked_job_id`: The tracked job ID — `trackedJobs[].id` from tracker_list output, distinct from `trackable.id` and `job.id` (required) - `remind_at`: Reminder date/time in ISO 8601 format, e.g. "2025-03-15T10:00:00Z" (required, must be in the future) **Output:** Returns the updated tracked job with reminderAt field.

Tracks a job from jobs_search results in the user's job tracker, identified by its job_id. For a job found elsewhere on the open web (with a URL but no jobs_search job_id), tracker_add_external is the right tool instead. Fields: - `job_id`: the job ID from jobs_search results (required) - `status`: initial status (saved, applied, interviewing, offered, archived); defaults to "saved" - `sub_status`: sub-status within the main status: saved: interested, researching_company, preparing_application, ready_to_apply; applied: application_submitted, followed_up; interviewing: interview_scheduled, phone_screen, technical, onsite, final_round, pending_feedback; offered: negotiating, considering, offer_received, accepted; archived: ghosted, rejected_by_company, withdrawn_by_candidate, not_interested, employed_by_this_company, employed_by_another_company - `notes`: notes about the job Returns the tracked job with its details. Repeated saves return the existing tracked job. A job that was previously removed from the tracker is restored with its earlier status and notes.

Saves a job posting found anywhere on the open web into the user's tracker. For jobs that came from jobs_search results, tracker_add (which takes a job_id) is the right tool instead. A job seen elsewhere in the conversation needs no prior jobs_search call — its URL and details from the conversation are sufficient input. `url`, `company_name`, `title_name`, `location_name`, and `description` identify the posting and are the only required fields. Every structured fact field (salary, dates, employment type, education, experience) is optional: a fact the source does not state is simply omitted (or null), and FoundRole's own extractors derive missing salary, employment, work-arrangement, education, experience, skills, benefits, and bonuses from the description. A save never waits on facts the source did not provide. The optional `client_extraction` object carries evidence-backed skills, technology, benefits, bonuses, seniority, industry, management, clearance, visa, and remote-scope labels when source excerpts for them exist; FoundRole validates and stores those labels separately. Fields: - `url`: the job posting's direct URL (required; not a company homepage) - `company_name`: company name (required) - `title_name`: job title (required) - `location_name`: location, e.g. "New York, NY" (required) - `description`: the posting's description from the source result; a short summary is acceptable (required) - `salary_min_value` / `salary_max_value`: salary range bounds (numbers) - `salary_value`: a single salary figure when there is no range (number) - `posted_at`: ISO 8601 posting date - `salary_currency`: ISO 4217 currency code - `salary_type`: one of year, month, week, day, hour - `employment_type`: array of full_time, part_time, contractor, temporary, intern, volunteer, per_diem, other - `work_location_type`: one of on_site, remote, hybrid - `education_requirements`: array of no_requirements, high_school, associate_degree, bachelor_degree, professional_certificate, pos…

Lists the user's tracked jobs with optional filtering and pagination. **Input:** - `status`: Filter by status (saved, applied, interviewing, offered, archived) - `limit`: Number of results per page (default 20, max 50) - `offset`: Number of results to skip (default 0) **Output:** Returns a list of tracked jobs grouped by status with pagination info. Each response includes a system_instruction describing how to present the results for the current client.

Removes a job from the user's job tracker. **Input:** - `tracked_job_id`: The tracked job ID — `trackedJobs[].id` from tracker_list output, distinct from `trackable.id` and `job.id` (required) **Output:** Confirms the job was removed from tracking.

Updates details of a tracked job (notes, deadline, salary, tags). **Input:** - `tracked_job_id`: The tracked job ID — `trackedJobs[].id` from tracker_list output, distinct from `trackable.id` and `job.id` (required) - `notes`: Updated notes - `deadline`: Deadline date (ISO 8601 format) - `salary_offered`: Salary amount - `salary_offered_type`: Salary type: year, month, week, day, hour - `tags`: Comma-separated tags (e.g., "remote,startup,tech") - `reminder_at`: Reminder date/time in ISO 8601 format, e.g. "2025-03-15T10:00:00Z" (must be in the future, or empty to clear) **Output:** Returns the updated tracked job.

Updates the status of a tracked job. **Input:** - `tracked_job_id`: The tracked job ID — `trackedJobs[].id` from tracker_list output, distinct from `trackable.id` and `job.id` (required) - `status`: New status: saved, applied, interviewing, offered, archived (required) - `sub_status`: Sub-status within the main status, valid only for that status (optional): saved: interested, researching_company, preparing_application, ready_to_apply; applied: application_submitted, followed_up; interviewing: interview_scheduled, phone_screen, technical, onsite, final_round, pending_feedback; offered: negotiating, considering, offer_received, accepted; archived: ghosted, rejected_by_company, withdrawn_by_candidate, not_interested, employed_by_this_company, employed_by_another_company **Output:** Returns the updated tracked job.