Freelance Proposal Engine

Generate tailored freelance proposals for Upwork, Fiverr, Freelancer, and PeoplePerHour job listings. Use when writing proposals, bidding on gigs, or responding to client job posts.

Security Scan

The skill's instructions, resource needs, and behavior are coherent with its stated purpose of generating tailored freelance proposals and do not ask for unrelated credentials or installs.

Purpose & Capability

The skill asks only to read job descriptions (paste, URL, or file) and fetch public listing pages; those capabilities match the stated purpose. The allowed tools (Read, Write, WebFetch, WebSearch) are appropriate. Note: inclusion of Bash/Grep/Glob is broader than strictly necessary for proposal text generation and may be excessive but is explainable (reading local files, formatting, simple text processing).

Instruction Scope

SKILL.md stays on-topic (analyze listing, draft proposal, recommend pricing, provide platform tips). It explicitly allows reading a provided file or fetching a listing URL. It does not instruct the agent to access unrelated system state or secret env vars. However the header permits powerful tools (Bash, Read, Write, Glob, Grep) which — if the runtime grants them — could access arbitrary local files or run shell commands; the instructions themselves do not justify broad shell access beyond reading a supplied job file.

Install Mechanism

No install spec and no code files (instruction-only), which is lowest-risk: nothing is downloaded or written to disk by an installer.

Credentials

The skill requires no environment variables, credentials, or external service tokens; requested capabilities are proportional to the task.

Persistence & Privilege

always is false and the skill does not request persistent system-wide changes or modify other skills. Autonomous invocation is allowed by default but is not combined with broad credentials or other red flags here.

Assessment

This skill appears coherent and low-risk: it only needs the job text or URL and produces proposals. Before enabling, check what runtime permissions the agent will actually grant this skill (especially Bash / file Read/Write and WebFetch): if possible, restrict shell/file access to only the directories you intend to share, and avoid providing sensitive local files. Test the skill with non-sensitive sample job listings first, and review any outgoing web requests or files it writes. If you don't need shell access, prefer a runtime configuration that disables Bash for this skill.

Generate high-converting freelance proposals tailored to specific job listings. This skill analyzes client needs, identifies pain points, and crafts proposals that win work.

How to Use

Provide the job listing in one of these ways:

• Paste the full job description as $ARGUMENTS
• Provide a URL to the listing (Upwork, Fiverr, Freelancer, etc.)
• Provide a file path containing the job description

Proposal Generation Process

Follow these steps exactly:

Step 1: Analyze the Job Listing

Extract and identify:

• Client pain points: What problem are they trying to solve?
• Explicit requirements: Skills, deliverables, timeline mentioned
• Implicit requirements: What they need but didn't say (read between the lines)
• Budget signals: Fixed price vs hourly, budget range if stated
• Red flags: Unrealistic expectations, scope creep potential, low budget
• Client experience level: First-time poster vs experienced buyer (review count, hire rate if visible)
• Keywords: Technical terms and buzzwords the client uses (mirror these back)

Step 2: Determine Proposal Strategy

Based on analysis, select the approach:

Client Type	Strategy
First-time buyer	Reassure, explain process, offer milestone-based payment
Experienced buyer	Be concise, lead with results, reference similar past work
Technical client	Use precise technical language, skip fluff
Non-technical client	Translate tech into business outcomes
Urgent project	Lead with availability and fast turnaround
Budget-conscious	Emphasize value, suggest MVP/phased approach

Step 3: Generate the Proposal

Use this structure:

**Opening Hook** (1-2 sentences)
- Reference a SPECIFIC detail from their listing (proves you read it)
- Connect it to a result you've delivered before
- Never start with "I" or "My name is" or "I'm a"

**Understanding Their Problem** (2-3 sentences)
- Restate their problem in your own words
- Show you understand the WHY behind the request
- Mention one thing they might not have considered

**Your Approach** (3-5 bullet points)
- Specific steps you'll take
- Tools/technologies you'll use
- Timeline for each step
- What they'll receive at each milestone

**Relevant Experience** (2-3 sentences)
- 1-2 specific similar projects (brief, results-focused)
- Quantified outcomes where possible ("increased conversions by 40%")
- If no exact match, draw parallels from adjacent experience

**Call to Action** (1-2 sentences)
- Suggest a specific next step (quick call, share examples, start immediately)
- Create mild urgency without being pushy
- Keep it conversational

Step 4: Pricing Recommendation

Based on the job analysis, suggest:

• Your recommended rate (based on market data and complexity)
• Rate justification (1 sentence)
• Alternative pricing: If the budget seems low, suggest a phased approach or reduced scope

Use these market rate guidelines:

Service	Beginner	Mid-Level	Expert
Web Development	$25-40/hr	$50-100/hr	$100-200/hr
Content Writing	$0.05-0.10/word	$0.10-0.25/word	$0.25-1.00/word
SEO	$30-50/hr	$75-150/hr	$150-300/hr
Web Scraping	$20-40/hr	$50-100/hr	$100-200/hr
Design	$25-50/hr	$50-100/hr	$100-250/hr
Data Analysis	$30-50/hr	$60-120/hr	$120-250/hr
Email Marketing	$25-40/hr	$50-100/hr	$100-200/hr
Social Media	$20-35/hr	$40-80/hr	$80-150/hr

Step 5: Output

Generate the proposal in a clean, copy-paste-ready format. Also provide:

• Platform-specific tips (e.g., Upwork: keep under 300 words, Fiverr: focus on deliverables)
• Questions to ask the client (2-3 clarifying questions that show expertise)
• Follow-up message template (for 48hrs after if no response)

Proposal Quality Rules

1. Never be generic. Every sentence must reference something specific from the listing.
2. Never oversell. Confidence without arrogance.
3. Never lie about experience. If you lack direct experience, say "I haven't done X specifically, but I've done Y which involves the same skills."
4. Keep it scannable. Clients review 20-50 proposals. Use short paragraphs and bullet points.
5. Word count: 150-300 words for simple jobs, 300-500 for complex ones. Never more.
6. No templates phrases: Ban "I am writing to express my interest", "I am confident that", "I look forward to hearing from you", "Dear Sir/Madam", "I have X years of experience."

Example

Job listing: "Need someone to scrape product data from 5 e-commerce sites. Need product name, price, description, images. CSV output. ~500 products per site."

Generated proposal:

Scraping 2,500 products across 5 e-commerce sites with clean CSV output — I've done this exact type of project multiple times.

Here's how I'd handle this:

• Day 1: Build scrapers for all 5 sites using Python + Playwright (handles JavaScript-rendered pages that simpler tools miss)
• Day 2: Run extraction, clean and normalize the data (consistent formatting across all 5 sources)
• Day 3: Deliver final CSVs with columns for product name, price, description, and image URLs

A few things that'll save you headaches: I'll handle pagination automatically, add retry logic for flaky pages, and deduplicate any products that appear in multiple categories.

Last month I scraped 15,000+ SKUs from three competitor sites for a retail client — delivered in 48 hours with 99.7% accuracy.

Quick questions: Are any of these sites behind a login? And do you need this as a one-time scrape or recurring?

I can start today if the details check out.