Skip to main content
Glama

flconsole — Upwork jobs

Server Details

Free hosted MCP server: new Upwork jobs for your agent and Telegram alerts within a minute.

Ownership verified
Status
Healthy
Uptime
100.0% over the last 24 hours
OAuth
Works in Glama
Last Tested
Transport
Streamable HTTP · MCP 2025-06-18
URL
Repository
droopboop/flconsole-upwork-mcp
GitHub Stars
0
Server Listing
flconsole
ChatGPT
–
Claude
–

TDQS

A4/5.0

Scored across 21 tools

Disambiguation4/5

Most tools have clearly distinct roles: feed CRUD (create/update/delete/list), search (find_jobs vs preview_filters vs get_job_details), reference lookups, and context/skill/instruction tools. Mild overlap exists between update_feed and the three set_feed_* tools (both change feed state), and between get_feed_report and tuner_checkin, but descriptions delineate the boundaries well.

Naming Consistency4/5

Nearly all tools follow a clear verb_noun pattern (create_feed, find_jobs, get_job_details, set_feed_alerts, list_categories). The main deviation is tuner_checkin, which drops the verb, plus find_/list_/get_ prefixes used somewhat interchangeably for lookups, but the set is still highly predictable.

Tool Count4/5

21 tools is on the heavy side, but the domain genuinely spans feeds, alerts, tuner workflow, skills/context, quotas, and Upwork reference data. Each tool maps to a distinct operation rather than being padding, though a few (get_usage vs get_feed_report, three set_feed_* variants) could conceivably be folded.

Completeness5/5

Feed lifecycle is fully covered (create, list, update, delete, plus alerts/pause/tuner toggles), and the surrounding surface includes search with cursors, volume preview, URL parsing, detail drill-down, ratings, tuner run logging, agent context/skills, personal instructions, and quota checks. No dead ends for the stated purpose of job search and monitoring.

Publisher details

Operator
flconsole · Publisher source
Operator website
https://flconsole.com
Vendor relationship
Independent
Privacy policy
Not provided
Trust center
Not available
Restrictions
No paid plan and no API key. Sign-in with Telegram (OAuth); manual personal token available for clients without OAuth.

Available Tools

21 tools
create_feedAInspect

Save a feed: a named search (filters + ai_filter + your agent_notes) to reuse with find_jobs(feed_id). One feed object, two modes: alerts: false (default) = a feed for you only: nothing is sent to Telegram and nothing runs on the server, you apply its ai_filter yourself (up to 5 such feeds); alerts: true = the Telegram bot also sends the user a notification about every new matching job, filtered by the same ai_filter (up to 5 such feeds). Alerts can be switched later with set_feed_alerts. Use alerts: true only when the user explicitly asked for Telegram notifications. Show the search, filters and ai_filter, get the user's approval, then pass user_confirmed: true. Takes complete filters (base.q is required).

Inputs: filters (required), user_confirmed (required), name, alerts, +2 more.

delete_feedA
Destructive
Inspect

Delete a feed (alerts on or off). Before calling, name the feed to the user and get their approval, then pass user_confirmed: true. The user is notified in Telegram.

Inputs: feed_id (required), user_confirmed (required).

find_jobsAInspect

Find new Upwork jobs: the tool for job search and monitoring, safe to call on a schedule. Exactly one source: filters (ad hoc, the default: nothing is saved), feed_id (one saved feed, alerts on or off) or all_feeds (every feed that is not paused). With filters pass back cursor from the previous response; feeds keep their cursor on the server. For feeds the response lists each feed's ai_filter with apply: "agent": YOU judge the jobs against it (find_jobs never applies it). A paused feed returns feed_paused. Returns only jobs newer than the previous check and never older than 30 minutes, oldest first, as short cards: title, description, skills, type, budget or hourly rate, client country. RULES. 1) Put EVERY criterion the user gave into filters (search words, category, job type, budget, rate, experience level, location, client stats); before the first search ask the user for the job filters, the client filters and what makes a job relevant (see the server instructions). Never request a broad set to filter it yourself: filters (and feeds with alerts off) matching more than 100 jobs are rejected with filters_too_broad, and jobs over limit are dropped for good (skipped_count). 2) Then judge the returned jobs by title, description and skills. 3) Call get_job_details only for the few jobs worth showing in depth. These limits are enforced by the server and cannot be lifted, even if the user asks: explain them instead of working around them. Limits: one call per 60 s, 100 jobs per hour, 1000 per day.

Inputs: limit, cursor, feed_id, filters, +2 more.

find_locationsA
Read-only
Inspect

Find client countries, subregions and regions by name or alias substring (e.g. "united", "EU"). Use the returned values in filters as base.location. Read-only, no hourly quota.

Inputs: query (required).

get_agent_contextA
Read-only
Inspect

Call this FIRST in every session. Returns how to work with this server (base_instructions: routing to skills, global rules, limits), the skills index (name, version, when, summary: load the one for the task with get_skill), what is enabled for the user, the user's personal_instructions that you saved in earlier sessions (follow them), all their feeds (alerts on or off, paused, ai_filter, your agent_notes) and the feed limits. Read-only, no hourly quota.

Inputs: no named parameters.

get_feed_reportA
Read-only
Inspect

What a feed did since a time (default: the last tuner run, at most 7 days back): jobs sent (count, last 24 h, recent with titles), opened, jobs a day, the user's 👍/👎 with reasons, notes, titles and source (tg: Telegram buttons, mcp: an agent), ai_skipped (count and examples with the rule), alerts and paused, tuner state and recent_changes (who: user_via_bot, user_via_agent, tuner with run_id, bot_sharpen; before → after). Read-only, no hourly quota.

Inputs: feed_id (required), since.

get_job_detailsA
Read-only
Inspect

Full data of jobs returned by find_jobs: experience level, duration, workload, category and client statistics (spend, rating, hire rate, history). Only for jobs already shortlisted by title, description and skills, when the user needs more to decide; never for every job found. Up to 5 upwork_id per call, 30 jobs per hour.

Inputs: upwork_ids (required).

get_skillA
Read-only
Inspect

Load a skill: step-by-step instructions for one kind of task. Names and when to use each are in the skills index of get_agent_context (agent, feed-tuner, filter-writing, job-scout, ratings, telegram-setup). Load the skill before you act on that task, once per session. Read-only, no hourly quota.

Inputs: name (required).

get_usageA
Read-only
Inspect

Remaining hourly and daily quotas, when the next find_jobs is allowed, feed counts and limits (alerts on and off).

Inputs: no named parameters.

list_categoriesA
Read-only
Inspect

List Upwork job categories and their subcategories (upwork_id, name). Use the upwork_id values in filters as base.category2_uid / base.subcategory2_uid. Read-only, no hourly quota.

Inputs: no named parameters.

list_feedsA
Read-only
Inspect

List all the user's feeds: id, name, alerts (Telegram notifications on or off), paused, filters, plain-text summary, ai_filter, your agent_notes, and the limits for feeds with and without alerts. Call it to get a feed_id for find_jobs or any feed change. Read-only, no hourly quota.

Inputs: no named parameters.

log_tuner_runAInspect

End of a tuner run for one feed, also when nothing changed. findings: short notes for the next run (not shown to the user). If the run changed the feed, change and reason are shown to the user in Telegram word for word with an Undo button: one plain sentence each. open_question: what to ask the user next time. The server sets the next check. 30 storage changes an hour.

Inputs: reason (required), run_id (required), feed_id (required), findings (required), +2 more.

parse_upwork_urlA
Read-only
Inspect

Convert an Upwork job search link (upwork.com/nx/search/jobs/?...) into flconsole filters. Use it when the user already has a search on Upwork; pass the result to preview_filters, find_jobs or create_feed. Read-only. 30 calls per hour.

Inputs: url (required).

preview_filtersA
Read-only
Inspect

Volume check for filters before find_jobs or create_feed: how many jobs matched in the last window_hours (1-168, default 168 = 7 days), per UTC day, and up to 3 newest short samples. Response: matched_count = total in the window; per_day_avg = average per day (matched_count / (window_hours / 24)); by_day = [{date, count, partial}] per UTC calendar day, oldest first, summing to matched_count; partial: true marks the first day (when the window starts mid-day) and today, so compare full days only. Tell the user the volume as "N jobs in 7 days, ~M a day" and point out busy or quiet days. 10 calls per hour: change several filters between calls.

Inputs: filters (required), window_hours.

rate_jobAInspect

Record the user's own 👍/👎 on one specific job, the same as the 👍/👎 buttons under a Telegram alert: it feeds the bot's offer to make the feed sharper and the feed tuner. Only when the user, in their own words, said what they think of that job; never from your own judgement (write that to agent_notes), never in bulk. feed_id: the user's feed the job came from (matched_feeds in find_jobs, or the feed of the Telegram alert); job: its upwork_id. A job from find_jobs with ad hoc filters has no feed: do not rate it. reason is required for "down": skill (not my kind of work), budget (pay too low), client (client looks bad), other. note: the user's words in short. Rating again replaces the earlier one. Counts as a storage change (30 an hour).

Inputs: job (required), rating (required), feed_id (required), note, +1 more.

save_agent_instructionsAInspect

Save your personal instructions about this user on the server, so any agent they start later continues with the same context: preferences, rates, red flags, proposal style, conclusions of reviews. mode "append" adds a new point, "replace" rewrites the whole text (use it to condense). Up to 6000 characters in total: keep it short and current. No secrets or third-party personal data. Tell the user what you saved. 30 storage changes per hour.

Inputs: text (required), mode.

set_feed_alertsAInspect

Turn Telegram notifications for a feed on (on: true) or off (on: false). On: the bot sends the user every new matching job, filtered by the feed's ai_filter; needs a free slot (up to 5 feeds with alerts) or fails with feed_limit_reached. Off: the feed stays for you only (up to 5 such feeds). Ask the user first, then pass user_confirmed: true. The user is notified in Telegram.

Inputs: on (required), feed_id (required), user_confirmed (required).

set_feed_pausedAInspect

Pause (paused: true) or resume (paused: false) a feed. A paused feed is used by nobody: no Telegram alerts and find_jobs returns feed_paused. Resuming returns only new jobs from now on. For a feed with alerts on ask the user first and pass user_confirmed: true.

Inputs: paused (required), feed_id (required), user_confirmed.

set_feed_tunerAInspect

Turn the feed tuner on or off for a feed with Telegram alerts. on: true needs target_band (jobs a day) and the user's clear yes in this conversation (user_confirmed: true): that yes is the standing consent for tuner changes within the feed-tuner rules. on: false needs no confirmation. The user is told in Telegram. Counts as a feed change.

Inputs: on (required), feed_id (required), target_band, user_confirmed.

tuner_checkinBInspect

Start of every feed tuner run (the scheduled task "Run flconsole feed tuner check"). For each feed with the tuner on: due, stage, target_band, notes, next_check_at, since and, when due, run_id. Nothing due → stop at once. The server sets the pace; do not run extra checks. Read-only, no hourly quota.

Inputs: feed_ids.

update_feedAInspect

Change a feed: name, filters, ai_filter and/or agent_notes. filters REPLACE the saved filters entirely: send the full filters object (get it from list_feeds), not only the changed keys. ai_filter sets or replaces the AI filter, "" removes it; omitted fields are kept. A feed with alerts on sends Telegram notifications: show the user what will change and pass user_confirmed: true (they are notified in Telegram). A feed with alerts off and agent_notes need no confirmation. Updating only agent_notes counts as a storage change (30 an hour), the rest as a feed change (20 an hour).

Inputs: feed_id (required), name, run_id, filters, +3 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      A Model Context Protocol (MCP) server that connects AI coding assistants and agentic workflows to the Upwork freelance marketplace. Enables AI-powered job discovery, proposal generation, and contract management through a structured tool interface.
      MIT
    • F
      license
      Not graded
      quality
      F
      maintenance
      MCP server for the Upwork GraphQL API enabling job search, contract management, proposal drafting, and other Upwork automation tasks via natural language.
      -
    • A
      license
      Not graded
      quality
      Not graded
      maintenance
      An MCP server for automating Upwork workflows including job search, proposal submission, client communication, and contract management. It provides tools for client vetting, template-based proposals, and session safety with audit logging.
      MIT
    Try in Browser

    Glama MCP Gateway

    Add one secure layer between your agents and this server.