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 agentmods add skills/megamen32/lasthumancommit/real-use-testingnpx skills add megamen32/LastHumanCommit --skill real-use-testinggit clone --depth 1 https://github.com/megamen32/LastHumanCommitWhat 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 | $0.00048 | $0.00607 |
| Opus 5 | $0.00024 | $0.00303 |
| Sonnet 5 | $0.00010 | $0.00121 |
| Haiku 4.5 | $0.00005 | $0.00061 |
Grade A, and why
real-use-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 2d 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Real Use Testing
Use when a user-facing claim needs real-surface proof. This is the final gate before "done"; test files never substitute for it. The journey executed is the minimal path's shortest real canary, not an invented broader tour.
Interaction tool ladder
Climb from the best available rung. Never claim an action is impossible before trying the last one.
- Accessibility tree first, as far as possible. Take an a11y snapshot
(
take_snapshot, accessibility tree, screen-reader structure). Target elements by role + accessible name + stable ref. This is the cheapest and most reliable targeting and usually suffices for the whole journey. - Improved agent/browser MCPs when available and suitable:
browserclaw— a11y snapshot + numbered ref targeting, agent-first;touchpoint— accessibility tree across desktop apps, not just browsers;agent-browser— high-level agent browser automation;playwright-mcp/chrome-devtools-mcp— pages, network, console, screenshots, performance.
- CDP / Playwright scripting when no MCP fits the surface or precise network/console evidence is required.
- Raw XY coordinate clicks + keyboard as the last resort. Some surfaces respond only to real coordinates. ALWAYS try this rung before reporting a UI action as impossible, and verify every coordinate hit with a fresh snapshot or screenshot.
Ladder discipline: after any navigation or state change, re-snapshot before the next action; a stale ref means re-snapshot, never guess; confirm each decisive action by observed state, not by command success.
Procedure
- Read the accepted claim, proof strength, target surface, allowed actions/test data, and stop conditions.
- Run the shortest actual user journey end-to-end using the ladder, without source-based shortcuts.
- Capture evidence appropriate to the claim and surface: a decisive screenshot for browser/UI outcomes, the exact command and output for CLI outcomes.
- Report
PASS,CHANGES_REQUIRED, orSTOP_MISSING_REAL_SURFACEwith the exact journey, observed result, evidence path/reference, accepted claim, and smallest repair.
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.
- 2d ago First seen · 58 lines · 48 tokens per session scan A 8f6e305e396f
real-use-testing is a skill published in the GitHub repository megamen32/LastHumanCommit (2 stars, last pushed 2d ago), licensed MIT. It adds 48 tokens to every session and 607 once invoked, about $0.0002 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
use-agent-browser-for-airi
Test AIRI display-model imports with agent-browser across stage-tamagotchi Electron, stage-web, and stage-pocket mobile web layouts. Use when uploading and verifying contributor-supplied Live2D ZIP, VRM, or MMD ZIP/PMX/PMD files through AIRI's model selector, including onboarding bypass, format-specific import…
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
peekaboo
Capture and automate macOS UI with the Peekaboo CLI.
repro-admin
Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.
browser_cdp
通过 Chrome DevTools Protocol (CDP) 连接已运行的浏览器，或扫描本机 CDP 端口，用于远程调试与多工具共享浏览器实例。.