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/threat-vector-security/guardian-agent/browser-session-defensenpx skills add Threat-Vector-Security/guardian-agent --skill browser-session-defensegit clone --depth 1 https://github.com/Threat-Vector-Security/guardian-agentWrote 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/threat-vector-security/guardian-agent/browser-session-defense)<a href="https://agentmods.dev/skills/threat-vector-security/guardian-agent/browser-session-defense"><img src="https://agentmods.dev/badge/skills/threat-vector-security/guardian-agent/browser-session-defense.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.00418 |
| Opus 5 | $0.00000 | $0.00209 |
| Sonnet 5 | $0.00000 | $0.00084 |
| Haiku 4.5 | $0.00000 | $0.00042 |
Grade A, and why
browser-session-defense 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 6d 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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Browser Session Defense
Use this when the work is about security boundaries around Guardian-managed browser tools rather than generic web UI testing.
Core Boundary
- GuardianAgent can directly control and reason about browsing that happens through its managed browser tools.
- GuardianAgent does not get rich page-level visibility into the user's normal browser sessions unless additional components exist outside the current runtime.
Workflow
- Clarify whether the browsing is:
- Guardian-managed
- user-driven in a normal browser
- For Guardian-managed browsing:
- prefer read-only page understanding first
- use the lowest-power browser tool that answers the question
- treat uploads, storage-state operations, and script execution as higher-risk
- For user-driven browsing:
- explain the visibility boundary clearly
- fall back to host and network signals when appropriate
Tooling Guidance
- Read-only page work: prefer
browser_read,browser_links, andbrowser_extract. - Interactive browser work: use
browser_stateplusbrowser_act, or Playwright tools directly when you need lower-level control. - Do not use
browser_run_code. - Treat
browser_evaluate, file uploads, and storage-state operations as approval-worthy actions.
Boundaries
- Use
webapp-testingfor functional UI verification and bug reproduction. - Use
host-firewall-defensewhen the only available evidence is indirect host or network drift around browser activity. - Use
security-mode-escalationwhen browser risk needs to be translated intomonitor,guarded, orlockdownrecommendations.
Gotchas
- Do not imply GuardianAgent can see arbitrary tabs or page contents in the user's own browser when it cannot.
- Do not default to the most powerful raw MCP tool for basic page reading.
- Do not recommend storage-state or upload actions as if they were harmless read-only inspection.
Read references/visibility-boundary.md when the user is confused about what GuardianAgent can and cannot observe in browser activity.
What ships with it
2 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.
- 6d ago First seen · 43 lines · 0 tokens per session scan A 548431bbd07d
browser-session-defense is a skill published in the GitHub repository Threat-Vector-Security/guardian-agent (11 stars, last pushed today), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 418 tokens. 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
browser
Web browser automation with AI-optimized snapshots for claude-flow agents.
browser
Browser automation via the agent-browser CLI. Use when the user needs to drive websites or Electron desktop apps — navigating, filling forms, clicking, screenshots, extracting data, testing web apps, visual UI checks, the Pi Dashboard's Electron shell, or the user's own logged-in browser (SSO/2FA sites). Triggers…
agentlas-browser
Use when the user types /agentlas-browser , /agentlas browser, or /hep-browser for browser-required tasks, rendering pages, form flows, or visual evidence.
playwright-cli
Skill "playwright-cli" from Insajin/autopus-adk, covering browser automation with playwright-cli, quick start, commands, core and navigation.
run-dashboard-e2e-local-changes
Run Playwright E2E (tests/e2e/) against the docker/ all-in-one harness so it reflects LOCAL code changes, not a stale cached image.
continuum-tools-mcp
Connect MCP servers (Stdio/SSE/StreamableHTTP) to a Continuum agent, configure tool filtering, set up tool-context capture/injection (e.g. sessionid), and read run artifacts (UI widgets, structured tool data). Invoke when the user asks "connect MCP", "filesystem tool", "remote API tool", "auto-capture sessionid"…