agentleFS
binay-pybinay-py/claude-skills/skills/freelance-proposal/SKILL.md

freelance-proposal

Skill1 starsChanged 6 days agoMITView on GitHub

Write Upwork proposals, Fiverr gig descriptions and cold client emails that get replies, built only from the freelancer's real work, then score them with proposal-lint (generic openers like "Dear Hiring Manager / I am writing to", AI-isms like leverage/seamless/delve, placeholders left in, I-focused pitch, no proof, no question, ignoring the job post, wrong length, weak 200-character preview). Use when the user pastes a job post, asks for a proposal, cover letter, gig description, or client outreach message.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the job post. Pull out: the real problem, stack/platform, deadline, budget, and anything unusual they mention (that's what proves you …

    line 12
  2. Distinctive:

    Write it in this shape (Upwork 80–200 words, Fiverr gig intro 40–160, cold email 50–150):

    line 14

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

freelance-proposal

Hard rule: never invent experience

Only use projects, numbers, links and skills the user actually has. Ask for, or read, a portfolio file (e.g. portfolio.md with project, link, stack and one result each). If nothing relevant exists, say so honestly and offer a small paid test task or a quick free sample instead of faking it. Invented experience gets accounts banned and loses clients.

Workflow

  1. Read the job post. Pull out: the real problem, stack/platform, deadline, budget, and anything unusual they mention (that's what proves you read it).
  2. Pick the one most relevant real project as proof (link + one concrete result).
  3. Write it in this shape (Upwork 80–200 words, Fiverr gig intro 40–160, cold email 50–150):
    • Line 1 (the preview, ~200 chars shown in the Upwork list): their problem or the outcome, in their words. Never a greeting, never "I am…".
    • Proof: "I built the same thing for X: <link>" plus one number if real.
    • Plan: 2–3 short steps specific to their post.
    • Logistics: timeline and price, if the post invites it.
    • One specific question, then a next step ("Can start tomorrow").
  4. Save it to a file and run the linter, with the job post for relevance:
    node scripts/proposal-lint.mjs proposal.txt --job job.txt [--platform upwork|fiverr|email]
    Fix every ✘ and ! item and re-run until it scores ≥ 85. Show the user the final text and the score. Never send it for them.
  5. Offer a second variant with a different hook (outcome-first vs problem-first) so they can choose.

Fiverr gig descriptions

Same rules, plus: first sentence = who it's for and the result ("Local clinics: a mobile site patients can book from in one tap"); then what's included per package, delivery time, what you need from the buyer, and 1–2 real portfolio links if Fiverr allows them in that field. No keyword stuffing.

Tone

Plain words, short sentences, contractions are fine. No "delve, leverage, seamless, robust, cutting-edge, tailored, elevate, top-notch, go above and beyond, perfect fit". No em-dash chains. Write like a busy expert texting a client, not a cover letter.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.