21 tools from the Jobsync MCP Server, categorised by risk level.
View the Jobsync policy →airtable_get_schema Return the field map this server writes to Airtable (the expected schema of the user's table). Use to diagnose schema-drift errors or to show the u... airtable_list_bases List all Airtable bases the PAT has access to. Useful during onboarding to help the user pick an existing base, or to confirm a newly created one. airtable_list_recent_jobs List recent job records from the user's Airtable base for dedup lookups. Returns the Apply Link, Company, Position Title, and Date for records crea... cache_is_seen Check whether a job URL has already been seen (local SQLite cache at ~/.jobsync/cache.db). Returns { seen: boolean, record?: {...} }. On a cache mi... classify_job_batch Composite classifier over an array of jobs. Runs US-location filter, title-keyword filter, and industry/tag/H1B/job-board detection in one call. Re... detect_industry_tags Deterministic classifier. Given a job's company name and title, returns the matched industry (from a fixed taxonomy), the company tier tags (FAANG+... fetch_ashby_jobs Fast-path: pull all current postings from Ashby boards via api.ashbyhq.com/posting-api/job-board. Slug is the company's Ashby identifier, e.g. 'ant... fetch_greenhouse_jobs Fast-path: pull all current postings from Greenhouse boards via the public JSON API (boards-api.greenhouse.io). Slug is the company's board identif... fetch_lever_jobs Fast-path: pull all current postings from Lever boards via api.lever.co/v0/postings. Slug is the company's Lever identifier, e.g. 'netflix' for job... filter_title_keywords Cheap pre-filter for a job title. Rejects excluded seniority/role keywords and requires at least one include keyword. Optionally enforces US-only l... filter_us_location Decide whether a location string looks non-US. Returns { likelyNonUS: boolean }. Use before deciding to keep or discard a job posting when usOnly m... jobsync_ping Sanity-check tool. Returns pong with server version. Use to verify the MCP server is connected. markdown_append_jobs Append processed jobs as rows to the markdown log at config.markdownPath (default ~/.jobsync/jobs.md). Alternative to Airtable when the user hasn't... profile_read Read the user's profile: skills.md, experience.md, projects.md, roles.json, and the raw resume text (if any). Use this at the start of every scrape... airtable_create_base Create a new Airtable base with the JobSync schema (Jobs table + all fields the server writes). Requires a PAT with the `schema.bases:write` scope ... airtable_upsert_job Create one or more job records in the user's Airtable base. Accepts an array of job objects matching the ProcessedJob shape (positionTitle, company... cache_mark_seen Record one or more jobs as seen in the local SQLite cache. Call this after successful `airtable_upsert_job` so the next run skips duplicates withou... profile_update_roles Update the user's role lists in roles.json. Pass any of `detected`, `custom`, `excluded` to replace that list, or `addCustom`/`addExcluded`/`remove... profile_write_file Write one of the profile markdown files (skills, experience, or projects). Used by the onboarding agent after extracting structured content from th... The Jobsync MCP server exposes 21 tools across 4 categories: Read, Write, Destructive, Execute.
Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the Jobsync server.
Jobsync tools are categorised as Read (14), Write (5), Destructive (1), Execute (1). Each category has a recommended default policy.
Deterministic policy on every MCP tool call. Per-identity grants. Full audit log.