Upwork

Skill mehtab78/skills/skills/upwork

Claude Code skills: freelance business pack and model-routing expert team that delegates to haiku/sonnet/opus/fable specialist subagents

Install

npx -y skills add mehtab78/skills --skill upwork

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use to win and manage freelance work on Upwork, writing personalized (non-templated) proposals, optimizing the profile/title/bio/rate, drafting client messages and negotiations, improving Job Success Score, and detecting scam or problem clients before you connect on one. Trigger whenever the user mentions Upwork proposals, bidding on a job, Connects, JSS, an Upwork client or contract, "help me reply to this Upwork message", or vetting a suspicious job post. The agent only drafts; the human reviews and sends. It never automates Upwork or touches account credentials.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.4 KB, as published. Nobody here has run it

Winning and running Upwork work

Upwork rewards freelancers who write like a human, respond fast, and keep contracts healthy. This skill helps with all of that, but within firm limits, because Upwork enforces its rules aggressively.

Non-negotiable rules

These aren't style preferences; breaking them risks the user's account:

The human sends, always. Draft proposals and messages, but the user reviews and submits them manually. Never auto-send.
No account access. Never log in, never touch credentials. The user copy-pastes what's needed.
No automation or scraping of Upwork actions or data.
Honest timelines. Don't promise delivery faster than is realistic.
On-platform payment only. Warn against any off-Upwork payment; it's a ToS violation and a ban risk.

Where the work lives

Task	Reference
Writing proposals that get replies	`references/proposals.md`
Profile, title, bio, rate	`references/profile.md`
Client messaging & negotiation	`references/communication.md`
Fees, JSS, contracts, Connects	`references/platform.md`
ToS boundaries & compliance	`references/legal.md`

Vetting a job before spending Connects

Flag any of these before the user commits; they predict wasted Connects or a bad contract:

Budget under ~$100 for genuinely complex work
"Need it ASAP" paired with a low budget
Client with no payment history or a weak rating
Requests for free "test" work
Wants to move to WhatsApp/email immediately
Vague scope ("build me an app")
Any payment outside the platform

The honest framing

This is productivity assistance, not automation. Everything the agent produces is a draft for the user to review, and the user is responsible for complying with Upwork's terms. Said plainly so neither side forgets it.