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 sinyu1012/proxypin-mcp-workbench --skill proxypin-capture-workflowgit clone --depth 1 https://github.com/sinyu1012/proxypin-mcp-workbenchWrote 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/sinyu1012/proxypin-mcp-workbench/proxypin-capture-workflow)<a href="https://agentmods.dev/skills/sinyu1012/proxypin-mcp-workbench/proxypin-capture-workflow"><img src="https://agentmods.dev/badge/skills/sinyu1012/proxypin-mcp-workbench/proxypin-capture-workflow/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/sinyu1012/proxypin-mcp-workbench/proxypin-capture-workflow"><img src="https://agentmods.dev/badge/skills/sinyu1012/proxypin-mcp-workbench/proxypin-capture-workflow.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00065 | $0.00680 |
| Opus 5 | $0.00032 | $0.00340 |
| Sonnet 5 | $0.00013 | $0.00136 |
| Haiku 4.5 | $0.00006 | $0.00068 |
Grade A, and why
proxypin-capture-workflow 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 11d 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ProxyPin Capture Workflow
Use the live ProxyPin instance and captured requests as the source of truth. Do not infer an endpoint from an old note when it is cheap to inspect the current capture.
Locate the live system
- The source checkout may live anywhere; resolve it from the user's workspace instead of assuming a personal absolute path.
- The customized desktop build commonly exposes an HTTP proxy and a loopback MCP server. Verify both from the running process, configuration, or
/health; do not assume ports are unchanged. - Preserve a dirty worktree. Before code changes, run
git status --shortand inspect overlapping diffs. - A running app, an open proxy port, captured CONNECT traffic, decrypted HTTPS, and a successful Mock match are separate verification levels. Report only the levels actually observed.
Route by task
- For phone connection, CA installation, HTTPS enablement, or “phone has no internet”, read references/capture-setup.md.
- For request discovery, MCP JSON-RPC, endpoint analysis, and request-detail inspection, read references/mcp-analysis.md.
- For complete data export, privacy-safe replay, validation, or Mock collection management, read references/export-and-mock.md.
Use scripts/proxypin_mcp.py for health checks and ordinary MCP calls. It redacts common credentials by default. Use --raw only when a local transformation genuinely needs exact values, and redirect raw output to a permission-restricted temporary file rather than echoing it into chat.
Shared boundaries
- Work only on traffic, accounts, devices, and applications the user owns or is authorized to test.
- Installing a local CA is authorized only for the user's device and ProxyPin instance. Never provide certificate-pinning bypass, app repackaging, Frida injection, jailbreak, or root instructions.
- “Analyze”, “inspect”, or “export” is read-only with respect to remote services. It does not authorize Mock rules, replays that mutate server state, account changes, or private endpoint probing.
- Mocking a paid or privileged state is acceptable only for the user's own application or an explicitly authorized test environment. Do not help bypass third-party entitlements.
- Treat cookies, Authorization headers, query tokens, account codes, device identifiers, child/family data, and captured bodies as sensitive. Do not print them in summaries, commits, screenshots, or persistent helper scripts.
- Prefer exact host-scoped HTTPS decryption. Do not enable blanket decryption when the task concerns a small set of domains.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 37 lines · 65 tokens per session scan A 188c7fd0a553
proxypin-capture-workflow is a skill published in the GitHub repository sinyu1012/proxypin-mcp-workbench (46 stars, last pushed 18d ago), licensed Apache-2.0. It adds 65 tokens to every session and 680 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 2026-08-30.
Other skills, from other repositories
flutter-mcp-toolkit-debug
Diagnose problems in a running Flutter app — read logs, evaluate Dart expressions, interpret error envelopes. Use when something broke.
firebase-crashlytics
Use when implementing crash reporting, capturing fatal/non-fatal errors, recording isolate/async exceptions, customizing reports, or uploading obfuscated symbols.
flutter-mcp-toolkit-inspect
Read state from a running Flutter app — semantic snapshot, view details, errors, screenshots, VM info. Use when you need to understand what the app is showing.
flutter-errors
Use when hitting layout errors (RenderFlex overflow, unbounded constraints, RenderBox not laid out), scroll errors, or setState-during-build errors.
flutter-mcp-toolkit-intentcall-migration
Migrate Flutter app code from removed MCPCallEntry to AgentCallEntry after the hard cut. Use when upgrading mcptoolkit, fixing compile errors after a major bump, or running flutter-mcp-toolkit migrate agent-entries.
flutter-mcp-toolkit-maintain-macos
Maintains fluttertestapp macOS showcase, native intentcall hooks (codegen, applinks invoke), and VM MCP validation. Use when editing macOS Runner, intentcallcodegen.sh, macOS dogfood, Screen Recording capture, or comparing macOS parity to web WebMCP.