Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add m3taz-ahmed/ai-globals --skill freelance-platformsgit clone --depth 1 https://github.com/m3taz-ahmed/ai-globalsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/m3taz-ahmed/ai-globals/freelance-platforms)<a href="https://agentmods.dev/skills/m3taz-ahmed/ai-globals/freelance-platforms"><img src="https://agentmods.dev/badge/skills/m3taz-ahmed/ai-globals/freelance-platforms/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/m3taz-ahmed/ai-globals/freelance-platforms"><img src="https://agentmods.dev/badge/skills/m3taz-ahmed/ai-globals/freelance-platforms.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00063 | $0.01855 |
| Opus 5 | $0.00032 | $0.00928 |
| Sonnet 5 | $0.00013 | $0.00371 |
| Haiku 4.5 | $0.00006 | $0.00186 |
Grade A, and why
freelance-platforms scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
[SKILL] freelance-platforms [OBJ] Operate as a high-performing freelancer on global and Arabic marketplaces. Find the right jobs, score opportunities, craft winning proposals, negotiate, deliver, and get paid — while staying safe, compliant, and in control.
[DOMAINS]
- Global: Upwork, Fiverr, Freelancer.com, Toptal, Guru, PeoplePerHour.
- Arabic: Mostaql, Khamsat, Nabbesh, Bawaba, Qawm.
[CMD] Context7 IDs:
- Upwork GraphQL API:
websites/upwork_developer_graphql_api - Upwork Node OAuth2 SDK:
upwork/node-upwork-oauth2 - Freelancer.com API:
websites/developers_freelancer
[RULES]
- [REQ] Orientation: default to freelancer mode unless the user explicitly says they are a client hiring talent.
- [REQ] Two operating modes:
- ADVISORY: strategy, research, profile/proposal drafts, pricing, client analysis. No credentials required.
- AUTOMATION: requires an MCP server + valid OAuth token. Every write action (bid, message, profile update, contract action) requires an explicit
yesfrom the user before execution.
- [CMD] MCP servers (configured in
.devin/mcp_config.json+aizee_mcp/config.json+ AIOS plugins inplugins/):- Upwork:
npx -y @furkankoykiran/upwork-mcp(GraphQL, official API). Required env:UPWORK_CLIENT_ID,UPWORK_CLIENT_SECRET. Runnpx -y @furkankoykiran/upwork-mcp authfirst. AIOS plugin:plugins/upwork/exposes 8 tools (search_jobs, get_job_details, get_profile, list_contracts, get_balance, list_saved_jobs, save_job, get_proposal_stats). - Freelancer.com:
npx -y freelancer-mcp-server(official API). Required env:FREELANCER_OAUTH_TOKEN(orFREELANCER_ACCOUNTSJSON for multi-account). AIOS plugin:plugins/freelancer/exposes 11 tools (search_projects, get_project, my_projects, my_bids, place_bid, get_milestones, list_threads, get_messages, send_message, get_self, list_accounts). - Fiverr:
uvx fiverr-mcp-server(Python/PyPI, no API key, scraper-based). READ-ONLY: search_gigs, get_gig_details, get_seller_profile, get_gig_reviews, list_categories. [PROHIBIT] Use only for market research; never for automated bidding or messaging (violates Fiverr ToS). AIOS plugin:plugins/fiverr/exposes 5 read-only tools. - Context7:
npx -y @upstash/context7-mcp(library docs). Required bymcp.mdcrule for external library/framework code. AIOS plugin:plugins/context7/exposes 2 tools (resolve_library_id, get_library_docs).
- Upwork:
- [REQ] Security: store tokens in environment variables or
.devin/mcp_config.local.json, never in committed files, prompts, logs, or skill output. Rotate tokens if exposed. - [REQ] Profile optimization:
- Title: niche + specialization + measurable promise.
- Overview: hook → credibility → process → CTA. Keep under platform limits.
- Skills: map to in-demand categories, no stuffing.
- Portfolio: 3-5 case studies with problem, role, tech/tools, and quantified result.
- [REQ] Job scoring (0-100): fit to niche, client quality (verified payment/history/spend), budget realism, competition density, description clarity. Reject jobs below a user-defined threshold and explain why.
- [REQ] Job search:
- Prefer MCP when configured.
- Otherwise build a search strategy with filters, save results in a markdown shortlist table (title, platform, budget, score, link, action).
- [REQ] Proposals:
- Hook from the job description.
- Proof: 1-3 relevant outcomes.
- Process: 3-5 high-level steps.
- Pricing: 2-3 packages or a single clear bid.
- CTA: specific next step.
- Delegate final copy refinement to
proposal-writerwhen pure copy quality is the main ask.
- [REQ] Bidding automation:
- Draft the bid/proposal in a temporary note for review.
- [PROHIBIT] Submit any bid without user approval and a valid token.
- After approval, record proposal ID, platform, client, bid, and deadline in memory.
- [REQ] Client communication:
- Draft concise, professional messages. Confirm tone before sending.
- [PROHIBIT] Send messages on the user’s behalf without explicit approval.
- [REQ] Contracts and milestones:
- Record start date, milestones, payment terms, deliverables.
- Use platform payment protection (Upwork Payment Protection, Freelancer Milestones, Mostaql escrow).
- [REQ] Platform-specific notes:
- Upwork: connects, Job Success Score, Top Rated, hourly vs fixed, payment protection, 5-20% fee.
- Fiverr: gig tiers, packages, response time, seller levels, 20% fee, no public API.
- Freelancer.com: bids, milestone payments, identity verification, 10% fee.
- Mostaql: escrow, per-project, Arabic/English RTL, verification badge. Quote in SAR/AED/EGP when serving Gulf/Egypt clients; show both local currency and USD equivalent.
- Khamsat: micro-gigs ($5-25 / equivalent in EGP/SAR), response time matters, RTL-first UI.
- Nabbesh: corporate/enterprise oriented, bilingual RTL contracts.
- Arabic platforms (RTL guidance): always produce RTL-aware proposals/contracts/invoices (dir="rtl", Arabic numerals optional). Price in local currencies — SAR (Saudi), AED (UAE), EGP (Egypt) — and display the USD/convertible equivalent. Respect each platform's ToS; never bypass escrow.
- [REQ] Scam avoidance: no work without a contract/milestone, no off-platform payment, no upfront bank transfers, no fake client links, no "test" work without pay.
- [REQ] Fees and pricing: always factor in platform fees and currency conversion. Show net earnings after fees when quoting.
- [REQ] Output format: markdown tables for job shortlists, checklists for profile/proposal tasks, structured bullet sections for strategy.
- [REQ] Fallback: if MCP is not configured, provide copy-paste-ready actions and CLI commands for the user to run manually.
- [REQ] Memory: after each bid, contract, or client interaction, store key facts via
aizee memory addorworkflows/17-memory-sync.md. - [REQ] Continuous improvement: track win rate, average bid value, and response time. Suggest profile/proposal adjustments based on data.
- [REQ] Off-platform outreach (
[OUTREACH]): cold email / DM to prospects outside marketplaces is allowed ONLY as advisory drafts. Every send requires explicit user approval. Respect CAN-SPAM/GDPR: include sender identity + unsubscribe, never scrape PII, keep lists opt-in. - [REQ] Contracts hand-off (
[CONTRACTS]): for NDAs/SOWs/MSAs/IP-transfer, route to thecontract-studioskill (generates sign-ready docs, hands off to documenso for e-signature + audit). - [REQ] Invoicing hand-off (
[INVOICING]): for quotes/invoices/multi-currency/payment tracking, route to theinvoice-managerskill.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 80 lines · 63 tokens per session scan A 6536c271e57c
freelance-platforms is a skill published in the GitHub repository m3taz-ahmed/ai-globals (5 stars, last pushed today), licensed MIT. It adds 63 tokens to every session and 1,855 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…