Upwork
v1.0.0Optimize Upwork profiles, write tailored proposals, identify client red flags, draft professional messages, and advise on pricing and client retention strate...
⭐ 2· 964·7 current·10 all-time
byIván@ivangdavila
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name and description (Upwork profile/proposal/communication help) match the contained markdown files (profile, proposals, communication, legal, platform). There are no requested binaries, environment variables, or config paths that would be unrelated to this purpose.
Instruction Scope
SKILL.md and supporting files explicitly restrict the agent to drafting/advice, mandate a human-in-the-loop, forbid logging into Upwork, forbid automation/scraping, and stay focused on communication, pricing, and ToS guidance. The instructions do not direct the agent to read system files, secrets, or send data to external endpoints.
Install Mechanism
This is an instruction-only skill with no install spec and no code files. That minimizes on-disk/execution risk — the skill consists of documentation the agent follows at runtime.
Credentials
The skill requires no environment variables, credentials, or config paths. It explicitly prohibits credential access and automation, so requested privileges are proportionate (none).
Persistence & Privilege
always:false (not forced into every agent run). The skill does not request persistent system presence or modification of other skills. It allows normal autonomous invocation by the agent platform (the default) but its rules require human review before actions that would affect user accounts.
Assessment
This skill appears coherent and focused on drafting and advising for Upwork. Before installing or using it: (1) do not paste Upwork credentials or confidential client NDA text into the agent — the skill forbids credential use but anything you paste could be included in prompts; (2) follow the human-in-the-loop rule — never auto-send proposals or messages; (3) verify the publisher/source if provenance matters to you (the registry owner is present but homepage/source is unknown); (4) watch for future updates that add install steps, env vars, or code — those would materially change the security posture; and (5) remember outputs are advisory and may be inaccurate or non-compliant with Upwork rules—always validate before acting.Like a lobster shell, security has layers — review code before you run it.
latestvk97aqvmy63n5rwv7dk10dhz32n815h7k
License
MIT-0
Free to use, modify, and redistribute. No attribution required.