Skip to main content
Glama

submit_proposal

Submit a proposal for an Upwork job by providing the job URL, cover letter, bid rate, and screening answers. Automates typing and submission to help you apply efficiently.

Instructions

Submit a proposal/bid for an Upwork job. Before calling this:

  1. Call get_job_details to get screening questions

  2. Craft a personalized cover letter that:

    • Opens with the client's specific problem (not "I am an expert in...")

    • Shows concrete n8n examples relevant to their use case

    • Mentions specific workflow patterns (HTTP Request, Webhook, Code node, etc.)

    • Ends with a clear CTA

  3. Answer ALL screening questions thoughtfully

The agent will automatically type and submit the proposal.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
job_urlYesFull Upwork job URL
cover_letterYesFull proposal text (cover letter)
bid_rateNoHourly rate or fixed price in USD
screening_answersNoAnswers to screening questions in order (from get_job_details)
boost_bidNoUse extra Connects to boost proposal. Default: false

Implementation Reference

  • The main handler function that executes the submit_proposal tool logic: navigates to the job, clicks 'Apply Now', sets bid rate, fills cover letter, answers screening questions, and submits the proposal via headless browser (Puppeteer/Playwright).
    export async function submitProposal(input: SubmitProposalInput): Promise<ProposalResult> {
      const page = await ensureLoggedIn();
      const bidRate = input.bid_rate ?? config.bid.default;
    
      try {
        console.error('[submitProposal] Navigating to job:', input.job_url);
        await page.goto(input.job_url, { waitUntil: 'domcontentloaded', timeout: 30000 });
        await humanDelay(2000, 4000);
    
        // Find and click "Apply Now" button
        const applySelectors = [
          '[data-test="apply-button"]',
          '[data-cy="apply-button"]',
          'a[href*="/proposals/new"]',
          'button:has-text("Apply Now")',
          'a:has-text("Apply Now")',
        ];
    
        let clicked = false;
        for (const sel of applySelectors) {
          const btn = page.locator(sel).first();
          if (await btn.isVisible({ timeout: 3000 }).catch(() => false)) {
            await btn.click();
            clicked = true;
            console.error('[submitProposal] Clicked apply button:', sel);
            break;
          }
        }
    
        if (!clicked) {
          return { success: false, message: 'Could not find Apply Now button. Job may be closed or already applied.' };
        }
    
        await humanDelay(2000, 4000);
        await page.waitForURL('**/proposals/**', { timeout: 15000 }).catch(() => {
          console.error('[submitProposal] URL did not change to /proposals/, continuing anyway');
        });
    
        // ─── Set bid rate ────────────────────────────────────────────────────────
        const hourlyRateInput = page.locator(
          '[data-test="rate-input"] input, input[name="rate"], #rate-input, input[placeholder*="rate"], input[placeholder*="Rate"]'
        ).first();
    
        if (await hourlyRateInput.isVisible({ timeout: 5000 }).catch(() => false)) {
          await hourlyRateInput.click({ clickCount: 3 });
          await hourlyRateInput.fill(bidRate.toString());
          console.error('[submitProposal] Set bid rate:', bidRate);
          await humanDelay(500, 1000);
        }
    
        // For fixed-price jobs — milestone / total price
        const fixedPriceInput = page.locator(
          'input[name="bid_amount"], input[placeholder*="Total price"], [data-test="bid-price"] input'
        ).first();
    
        if (await fixedPriceInput.isVisible({ timeout: 2000 }).catch(() => false)) {
          await fixedPriceInput.click({ clickCount: 3 });
          await fixedPriceInput.fill(bidRate.toString());
          console.error('[submitProposal] Set fixed price:', bidRate);
          await humanDelay(500, 1000);
        }
    
        // ─── Fill cover letter ───────────────────────────────────────────────────
        const coverLetterSelectors = [
          'textarea[name="cover_letter"]',
          '[data-test="cover-letter"] textarea',
          '[data-cy="cover-letter"] textarea',
          'textarea[placeholder*="cover"]',
          '.cover-letter textarea',
          '#cover_letter',
        ];
    
        let coverLetterFilled = false;
        for (const sel of coverLetterSelectors) {
          const textarea = page.locator(sel).first();
          if (await textarea.isVisible({ timeout: 3000 }).catch(() => false)) {
            await textarea.click();
            await textarea.fill('');
            // Type human-like
            await page.keyboard.type(input.cover_letter, { delay: Math.random() * 20 + 10 });
            coverLetterFilled = true;
            console.error('[submitProposal] Cover letter filled.');
            break;
          }
        }
    
        if (!coverLetterFilled) {
          console.error('[submitProposal] WARNING: Could not find cover letter textarea');
        }
    
        await humanDelay(1000, 2000);
    
        // ─── Answer screening questions ──────────────────────────────────────────
        if (input.screening_answers && input.screening_answers.length > 0) {
          const questionTextareas = await page.locator(
            '[data-test="additional-question"] textarea, [data-cy="qa-answer"] textarea, .screening-question textarea'
          ).all();
    
          for (let i = 0; i < Math.min(questionTextareas.length, input.screening_answers.length); i++) {
            const answer = input.screening_answers[i];
            if (answer) {
              await questionTextareas[i].click();
              await questionTextareas[i].fill('');
              await page.keyboard.type(answer, { delay: Math.random() * 20 + 10 });
              await humanDelay(500, 1000);
              console.error(`[submitProposal] Answered question ${i + 1}`);
            }
          }
        }
    
        await humanDelay(1500, 3000);
    
        // ─── Submit ──────────────────────────────────────────────────────────────
        const submitSelectors = [
          '[data-test="submit-proposal"]',
          '[data-cy="submit-proposal"]',
          'button[type="submit"]:has-text("Submit")',
          'button:has-text("Submit Proposal")',
          'button:has-text("Send Proposal")',
        ];
    
        let submitted = false;
        for (const sel of submitSelectors) {
          const btn = page.locator(sel).first();
          if (await btn.isVisible({ timeout: 3000 }).catch(() => false)) {
            console.error('[submitProposal] Clicking submit:', sel);
            await btn.click();
            submitted = true;
            break;
          }
        }
    
        if (!submitted) {
          return { success: false, message: 'Could not find Submit button. Proposal NOT submitted.' };
        }
    
        await humanDelay(3000, 5000);
    
        // Check for success
        const currentUrl = page.url();
        const successIndicators = [
          '[data-test="proposal-submitted"]',
          '[data-cy="proposal-submitted"]',
          '.proposal-submitted',
          ':has-text("Proposal Submitted")',
        ];
    
        for (const sel of successIndicators) {
          if (await page.locator(sel).isVisible({ timeout: 5000 }).catch(() => false)) {
            return {
              success: true,
              message: 'Proposal submitted successfully!',
              proposal_url: currentUrl,
            };
          }
        }
    
        // If URL changed to proposals list, likely succeeded
        if (currentUrl.includes('/proposals') && !currentUrl.includes('/new')) {
          return {
            success: true,
            message: 'Proposal submitted (redirected to proposals page).',
            proposal_url: currentUrl,
          };
        }
    
        return {
          success: true,
          message: 'Proposal likely submitted. Please verify manually.',
          proposal_url: currentUrl,
        };
      } catch (err) {
        const msg = err instanceof Error ? err.message : String(err);
        console.error('[submitProposal] Error:', msg);
        return { success: false, message: `Error submitting proposal: ${msg}` };
      } finally {
        await page.close();
      }
    }
  • Zod schema defining the input shape for submit_proposal: job_url, cover_letter, bid_rate, estimated_duration, screening_answers, boost_bid, plus the ProposalResult output type.
    export const SubmitProposalSchema = z.object({
      job_url: z.string().describe('Full URL of the Upwork job posting'),
      cover_letter: z
        .string()
        .describe(
          'The proposal/cover letter text. Should be personalized, mention the client\'s specific needs, and demonstrate n8n expertise.'
        ),
      bid_rate: z
        .number()
        .optional()
        .describe(
          `Hourly rate or fixed price bid in USD. Defaults to ${config.bid.default}. Range: ${config.bid.min}–${config.bid.max}.`
        ),
      estimated_duration: z
        .string()
        .optional()
        .describe('Estimated time to complete (for fixed-price jobs), e.g. "1 week", "3 days"'),
      screening_answers: z
        .array(z.string())
        .optional()
        .default([])
        .describe(
          'Answers to screening questions in order. Get questions first via get_job_details.'
        ),
      boost_bid: z
        .coerce.boolean()
        .optional()
        .default(false)
        .describe('Whether to use extra Connects to boost the proposal visibility'),
    });
    
    export type SubmitProposalInput = z.infer<typeof SubmitProposalSchema>;
  • src/index.ts:102-130 (registration)
    MCP tool registration in index.ts (the primary MCP server entry point) declaring the tool name 'submit_proposal' with its description and JSON Schema inputSchema.
      {
        name: 'submit_proposal',
        description: `Submit a proposal/bid for an Upwork job.
    Before calling this:
    1. Call get_job_details to get screening questions
    2. Craft a personalized cover letter that:
       - Opens with the client's specific problem (not "I am an expert in...")
       - Shows concrete n8n examples relevant to their use case
       - Mentions specific workflow patterns (HTTP Request, Webhook, Code node, etc.)
       - Ends with a clear CTA
    3. Answer ALL screening questions thoughtfully
    
    The agent will automatically type and submit the proposal.`,
        inputSchema: {
          type: 'object',
          properties: {
            job_url: { type: 'string', description: 'Full Upwork job URL' },
            cover_letter: { type: 'string', description: 'Full proposal text (cover letter)' },
            bid_rate: { type: ['number', 'string'], description: 'Hourly rate or fixed price in USD' },
            screening_answers: {
              type: 'array',
              items: { type: 'string' },
              description: 'Answers to screening questions in order (from get_job_details)',
            },
            boost_bid: { type: 'boolean', description: 'Use extra Connects to boost proposal. Default: false' },
          },
          required: ['job_url', 'cover_letter'],
        },
      },
  • src/index.ts:303-307 (registration)
    The switch-case dispatch in index.ts that parses incoming args with SubmitProposalSchema and calls submitProposal().
    case 'submit_proposal': {
      const input = SubmitProposalSchema.parse(args);
      result = await submitProposal(input);
      break;
    }
  • src/gateway.ts:100-114 (registration)
    Secondary MCP gateway registration in gateway.ts also declaring 'submit_proposal' with a simplified inputSchema for an alternative MCP endpoint.
    {
      name: 'submit_proposal',
      description: 'Submit a proposal for an Upwork job. Call get_job_details first to get screening questions.',
      inputSchema: {
        type: 'object',
        properties: {
          job_url:            { type: 'string' },
          cover_letter:       { type: 'string' },
          bid_rate:           { type: ['number', 'string'] },
          screening_answers:  { type: 'array', items: { type: 'string' } },
          boost_bid:          { type: 'boolean' },
        },
        required: ['job_url', 'cover_letter'],
      },
    },
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description must disclose behavioral traits. It states the agent will automatically type and submit, but fails to mention if the operation is destructive, reversible, or requires authentication. Missing details on side effects or rate limits.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is efficiently structured with a clear opening sentence, bullet points for prerequisites, and no redundant text. Slightly verbose but justified by the need for instruction.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with 5 parameters, no output schema, and no annotations, the description adequately explains preparation steps and parameter usage. However, it omits the return value (e.g., success/ failure) and error handling, leaving gaps for the agent to infer.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema descriptions cover all 5 parameters clearly, and the description adds value by referencing get_job_details for screening_answers and explaining bid_rate format. This exceeds the baseline expectation for high schema coverage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool submits a proposal/bid for an Upwork job, a specific verb+resource pair. It distinguishes itself from siblings like get_job_details (retrieval) and get_proposals (viewing), making the purpose unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides explicit prerequisites and steps before calling, such as calling get_job_details and crafting a cover letter. However, it lacks statements on when not to use the tool or alternatives, though the context is strong enough to guide proper usage.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/zcrossoverz/upwork-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server