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 mvanhorn/agentcookie --skill skillgit clone --depth 1 https://github.com/mvanhorn/agentcookieWrote 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/mvanhorn/agentcookie/skill)<a href="https://agentmods.dev/skills/mvanhorn/agentcookie/skill"><img src="https://agentmods.dev/badge/skills/mvanhorn/agentcookie/skill/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/mvanhorn/agentcookie/skill"><img src="https://agentmods.dev/badge/skills/mvanhorn/agentcookie/skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 6 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high YARA Match · line 3 YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).Fix: Remove the malware payload or compromised file entirely. Investigate how it entered the skill and audit all other artifacts for additional indicators of compromise.
- medium Privilege Escalation · line 51 Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
- medium Rogue Agent · line 82 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
- medium Privilege Escalation · line 90 Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
- medium Rogue Agent · line 159 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
- medium Rogue Agent · line 178 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.02259 |
| Opus 5.5 | $0.00025 | $0.00904 |
| Sonnet 5.5 | $0.00013 | $0.00452 |
| Haiku 4.5 | $0.00006 | $0.00226 |
Grade B, and why
agentcookie-install scanned grade B with 2 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 27d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
sudo mv agentcookie /usr/local/bin/ Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -LO https://github.com/mvanhorn/agentcookie/releases/download/v1.0.0/agentcookie_1.0.0_darwin_arm64.tar.gz How it starts
The opening of the file, as written. The whole thing — 242 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agentcookie install
You are helping the user install agentcookie on two machines that are both on the same Tailscale tailnet, then pair them, so that the sink's Chrome stays continuously in sync with the source's Chrome.
After install, the user does not touch agentcookie again. The source watches Chrome via fsnotify and pushes every cookie change to the sink within seconds. On Linux, the sink injects cookies via CDP into an already-running Chrome. On macOS, a LaunchAgent keeps the daemon running across reboots.
Inputs you need
- Which machine is the source (the Mac the user logs into Chrome on, usually their laptop).
- Which machine is the sink (where AI agents act - a Linux VM like Grok Bot, or a second Mac).
- Tailscale is up on both.
- On Linux: Chrome is running with
--remote-debugging-port=9223(or another port).
If any of these are missing, stop and ask.
Flow: Mac source to Linux sink (featured path)
Step 0: detect the lay of the land
Run on the current machine:
uname -s # Darwin = macOS, Linux = Linux
tailscale status 2>&1 | head -20
From the Tailscale status output, identify which host is the Mac and which is the Linux box.
Step 1: confirm source vs sink with the user
Use the platform's blocking question primitive. Phrase it concretely:
I see you're on
<current-hostname>. Looks like<other-hostname>(Tailscale IP100.x.y.z) is your other machine. Should I install agentcookie with<current-hostname>as the source (your logged-in Chrome) and<other-hostname>as the sink (where your agents run)?
Confirm before proceeding. If wrong, ask which is which.
Step 2: install on the Mac source
Install the binary if missing:
# Download from GitHub Releases
curl -LO https://github.com/mvanhorn/agentcookie/releases/download/v1.0.0/agentcookie_1.0.0_darwin_arm64.tar.gz
tar -xzf agentcookie_1.0.0_darwin_arm64.tar.gz
sudo mv agentcookie /usr/local/bin/
Or build from source:
go install github.com/mvanhorn/agentcookie/cmd/[email protected]
What ships with it
1 file 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.
- 27d ago Changed · +2 lines 1af38daca23a
- 1mo ago First seen · 240 lines · 63 tokens per session scan B bad7392e5498
agentcookie-install is a skill published in the GitHub repository mvanhorn/agentcookie (854 stars, last pushed 4d ago), licensed MIT. It adds 63 tokens to every session and 2,259 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
surf
Control Chrome browser via CLI for testing, automation, and debugging. Use when the user needs browser automation, screenshots, form filling, page inspection, network/CPU emulation, DevTools streaming, or AI queries via ChatGPT/Gemini/Perplexity/Grok/AI Studio.
ketch
Research skill for ketch — a fast stateless CLI for web search, OSS code search, curated library docs, page scraping, and site crawling; an optional MCP server exists for operators who want it, but the CLI is the primary interface. Use when a question needs live sources: 'research X', 'what are people saying about Y'…
page-shot
Take a headless screenshot of a URL the user gives.
browser-probe
Drive a browser with Playwright to read the page title.
aluvia
Use Aluvia when a cloud agent is in the wrong country for a site; when a page says not available in your region, this content is not available in your country, we don’t ship to your location, or other geo-restriction; when you need the browser to use an IP in a specific country (aluvia proxy-on --geo US; run aluvia…
docs-link-checker
Crawl a documentation site's own navigation and report broken links and empty pages.