Borrowing it
Nothing to install: this file belongs to enkhbold470/zangia-mcp-ai. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/enkhbold470/zangia-mcp-ai/main/.agents/skills/zangia-mcp-testing/SKILL.mdgit clone --depth 1 https://github.com/enkhbold470/zangia-mcp-aiWrote 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/enkhbold470/zangia-mcp-ai/zangia-mcp-testing)<a href="https://agentmods.dev/skills/enkhbold470/zangia-mcp-ai/zangia-mcp-testing"><img src="https://agentmods.dev/badge/skills/enkhbold470/zangia-mcp-ai/zangia-mcp-testing/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/enkhbold470/zangia-mcp-ai/zangia-mcp-testing"><img src="https://agentmods.dev/badge/skills/enkhbold470/zangia-mcp-ai/zangia-mcp-testing.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 | $0.00024 | $0.00521 |
| Opus 5.5 | $0.00010 | $0.00208 |
| Sonnet 5.5 | $0.00005 | $0.00104 |
| Haiku 4.5 | $0.00002 | $0.00052 |
Grade A, and why
zangia-mcp-testing 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 2mo 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.
What it actually says
Zangia AI + MCP Testing & Execution Skill
Install (any major MCP host)
# Claude Code
claude mcp add zangia -- npx -y zangia
# OpenAI Codex
codex mcp add zangia -- npx -y zangia
# Direct stdio
npx -y zangia
Claude Desktop / Cursor JSON:
{
"mcpServers": {
"zangia": {
"command": "npx",
"args": ["-y", "zangia"]
}
}
}
Local clone (dev):
bun run mcp:stdio
Honest data limits (read before evaluating location/salary)
Live crawl of zangia.mn homepage only exposes title + code + url.
| Field | Live crawl truth |
|---|---|
url |
Real https://www.zangia.mn/job/<code> |
district |
Inferred from title keywords only; else Unknown |
salary |
Unknown (max=0, salaryKnown=false) — open url |
| Downtown UB | Title landmarks: Сүхбаатар бизнес, Нарны зам, Зүүн дөрвөн зам, Маргад — not Сүхбаатар аймаг |
Cursor WebFetch is a separate Cursor browser tool, not part of this MCP. Agents should use it only when they need the full HTML posting page after MCP returns url.
Automated tests
bun run test
MCP tools
search_jobs(query?, category?, district?, isRemote?, minSalary?, limit?)get_job_details(jobId)— local IDs or livezangia-live-<code>/ bare codematch_candidate_cv(cvText, targetJobId?, minMatchScore?)filter_by_salary(minSalary?, maxSalary?, currency?)get_salary_benchmarks(category?)post_job(...)
Suggested downtown UB query
search_jobs({ district: "Sukhbaatar", limit: 20 })
Then open each url (or WebFetch that URL) for exact address/salary. Do not trust salary bands from list crawl.
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.
- 2mo ago First seen · 75 lines · 24 tokens per session scan A 7874e6a6435e
zangia-mcp-testing is a skill published in the GitHub repository enkhbold470/zangia-mcp-ai (2 stars, last pushed 2mo ago), licensed MIT. It adds 24 tokens to every session and 521 once invoked, about $0.0001 per session on Opus 5.5. 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
ae-ltd-skill-builder
Create new Claude Code Skills with proper YAML frontmatter, progressive disclosure structure, and complete directory organization. Use when building custom skills, generating skill templates, fixing naming issues, or understanding the Claude Skills specification.
ae-ltd-claude-template-switchboard
Use when creating or upgrading CLAUDE.md instructions and selecting the right template baseline for repo size, architecture, and team constraints. AE.LTD template decision workflow.
ae-ltd-context7-radar
Use when coding decisions depend on current library/framework APIs and you need version-accurate docs before implementation. AE.LTD workflow tuned for fast, verifiable doc grounding.
ae-ltd-mcp-foundry
Use when scaffolding or extending MCP servers, adding tools/resources/prompts, or hardening MCP tool schemas and handlers. AE.LTD workflow tuned for production-ready MCP delivery.
ae-ltd-mutation-gate
Use when unit tests look green but confidence is low and mutation testing is needed to expose weak assertions. AE.LTD workflow for targeted mutation runs and actionable test hardening.
ae-ltd-n8n-orbit
Use when converting plain-language automation goals into production-ready n8n workflow JSON with retries, error paths, and credential-safe placeholders. AE.LTD automation blueprint workflow.