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 lesterppo/hermes-gem-pw --skill hermes-gem-pwgit clone --depth 1 https://github.com/lesterppo/hermes-gem-pwWrote 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/lesterppo/hermes-gem-pw/hermes-gem-pw)<a href="https://agentmods.dev/skills/lesterppo/hermes-gem-pw/hermes-gem-pw"><img src="https://agentmods.dev/badge/skills/lesterppo/hermes-gem-pw/hermes-gem-pw/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/lesterppo/hermes-gem-pw/hermes-gem-pw"><img src="https://agentmods.dev/badge/skills/lesterppo/hermes-gem-pw/hermes-gem-pw.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 | $0.00039 | $0.01241 |
| Opus 5 | $0.00019 | $0.00620 |
| Sonnet 5 | $0.00008 | $0.00248 |
| Haiku 4.5 | $0.00004 | $0.00124 |
Grade A, and why
gem-pw 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
gem-pw — Gemini CLI via Browser Automation (No API Key)
Two CLIs, one repo: gem-pw (Gem operations) and gemini-web (non-Gem operations). Both drive a real Chromium browser via Playwright — the only backend that reliably reaches Google's browser-gated Gemini endpoints. Zero API keys, zero cookie extraction, zero external servers.
When to Use
- Chat with any Gemini Gem via URL/id (single or multi-turn)
- Gem CRUD: create, edit (name/instructions/model/knowledge), delete
- Image generation via Imagen (Upload & tools → Create image drawer)
- Deep research via
gemini-web research - Token-efficient code/design iteration via
--collabsurgical diffs - Direct chat without a Gem (fastest path, no Gem overhead)
gemini-webapireturns UNAUTHENTICATED — real browser bypasses this gate
Prerequisites
- Python 3.10+ and
pip install playwright aiohttp && playwright install chromium - One-time sign-in:
gem-pw-login(opens Chromium, profile persists at~/.gemini-cli/cr-profile/) - CDP path: a headed Chromium on
http://127.0.0.1:9224(default; override withGEM_PW_CDPor--cdp). Without CDP, falls back to self-launched headed Chromium.
Quick Reference
| Need | Command |
|---|---|
| Chat with a Gem | gem-pw <id> "prompt" |
| Multi-turn chat | gem-pw <id> -c sess.json "prompt" |
| Diff-based iteration | gem-pw --collab <id> -f file "instr" -o r.md |
| Apply the diff | python3 apply_gem_diff.py r.md file [--dry] |
| Create a Gem | gem-pw --create "Name" "Instructions" [-m model] [--knowledge-*] |
| Edit a Gem | gem-pw --edit <id> --name "New" --instructions "..." |
| Delete a Gem | gem-pw --delete <id> |
| Upload + ask | gem-pw --upload <id> -f file "question" |
| Image gen (Gem) | gem-pw --img <id> "description" |
| Image gen (main app) | gem-pw --img "description" |
| Direct chat (no Gem) | gemini-web chat "prompt" |
| Deep research | gemini-web research "topic" |
| Account info | gemini-web account |
Output Format
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 · 104 lines · 39 tokens per session scan A 055622f5b998
gem-pw is a skill published in the GitHub repository lesterppo/hermes-gem-pw (1 stars, last pushed 1mo ago), licensed MIT. It adds 39 tokens to every session and 1,241 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
webcmd-browser
Use when a live browser task requires Playwright interaction, authenticated handoff, visible UI verification, or ad-hoc page inspection.
BrowserBash Browser Automation
BrowserBash is a vendor-independent, natural-language browser automation CLI. Drive a real browser from plain-English objectives or committable Markdown tests, run on local Chrome, CDP/Playwright MCP, Browserbase, LambdaTest, or BrowserStack, and stream NDJSON results with CI exit codes — using free local Ollama…
auto-captcha-solver
Universal captcha auto-solver for Playwright browser automation. Detects hCaptcha and reCAPTCHA v2 on any page, solves via NopeCHA API, and injects tokens automatically. Drop-in CaptchaSolver class with autosolve(page) method.
rawtrace-debug-browser
Use RawTrace proactively to debug flaky or timing-sensitive browser behavior, including intermittent Playwright or Cypress failures, clicks with no request or unclear result, transient or asynchronous DOM changes, API timing, WebSocket ordering, downloads, authentication redirects, or cases where snapshots are…
qprobe
QUESTPIE Probe — dev testing CLI for AI coding agents. Start servers, test APIs, control browsers via agent-browser, record and replay regression tests with zero tokens. Use when testing web apps, starting dev servers, reading logs, debugging errors, making API calls, checking browser console/network, or composing…
instagram-workflow
A workflow for publishing image posts to Instagram, a social network for photos and videos, using browser automation.