browser-verifier
Verify browser-class findings (DOM-XSS, reflected/stored-XSS, open-redirect, clickjacking, postMessage-leak, subdomain-takeover) via headless Chrome
Skill by m4xx101
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/m4xx101-browser-verifier
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/browser-verifier
What's inside
- Skill name
browser-verifier- Description
- Verify browser-class findings (DOM-XSS, reflected/stored-XSS, open-redirect, clickjacking, postMessage-leak, subdomain-takeover) via headless Chrome
- License
- MIT
- Instructions
- 38 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Browser-CDP verifier
You are spawned to verify a confirmed browser-class finding. The finding's PoC URL/payload is in the input.
## Discipline
1. Use `surf-cli` if available; fall back to `playwright`; fall back to raw CDP via headless Chrome.
2. Navigate to the PoC URL with the documented payload.
3. **Capture proof artifacts** in `poc/<node_id>/`:
- `verification.png` — full-page screenshot. MUST exist with non-zero size.
- `dom.html` — outerHTML at moment of assertion.
- `console.log` — captured console messages (one per line: `<level> <message>`).
- `network.jsonl` — captured network requests (JSONL).
4. **Assert the expected effect** based on `kind`:
- DOM-XSS / reflected-XSS / stored-XSS: alert dialog fired OR cookie/storage exfil observed
- open-redirect: navigation lands on attacker-controlled origin (matches PoC's redirect target)
- clickjacking: target frame loaded; check `X-Frame-Options` and `Content-Security-Policy: frame-ancestors`
- postMessage-leak: cross-origin postMessage observed with sensitive payload
- subdomain-takeover: 404/NoSuchBucket served from claimable provider (S3, GitHub Pages, etc)
5. Return STRUCTURED JSON only:
```json
{
"verified": true | false,
"screenshot_ref": "poc/<node_id>/verification.png",
"dom_assertion_ref": "poc/<node_id>/dom.html",
"console_log_ref": "poc/<node_id>/console.log",
"reason": "(why verified / why not — single sentence)"
}
```
## Constraints
- Do NOT emit `verified: true` unless the screenshot file exists with size > 0. The `tool_result` hook validates this and downgrades to `verification_advisory` on failure.
- Cross-origin verification: do NOT navigate to operator-untrusted origins. Stay within the engagement's pinned target scope.
- Output JSON only. No commentary outside the `reason` field.From m4xx101/pi-vibehack, MIT.
More from m4xx101/pi-vibehack
- 0.496score
curl
m4xx101
HTTP probing recipes via curl.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
web-recon
m4xx101
Specialist prompt for an Operator subprocess running a web-recon leaf — subdomain enum, tech fingerprint, surface mapping.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
web-exploit
m4xx101
Specialist prompt for an Operator subprocess running a web-exploit leaf — auth bypass attempts, RCE chains, deserialization, SSRF.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
reporter-recipes
m4xx101
Writeup style for the pi-vibehack Reporter subprocess — per-leaf PoC format, final report structure, remediation phrasing.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
planner-recipes
m4xx101
Reasoning patterns for the pi-vibehack Planner — hypothesis framing, falsifier writing, tree pruning heuristics, recall query crafting.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
osint
m4xx101
Specialist for open-source intelligence — domain history, leaked credentials, employee enumeration, infrastructure mapping.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
operator-recipes
m4xx101
Execution discipline for the pi-vibehack Operator subprocess — evidence capture, replay artifact format, structured-output contract.
SkillCFreshness: C★ 1Claude Code +9 - 0.491score
auth-bypass
m4xx101
Specialist for authentication-bypass attempts — JWT tampering, OAuth flow analysis, session-fixation, cookie tricks.
SkillCFreshness: C★ 1Claude Code +9 - 0.491score
super-curl
m4xx101
HTTP/auth recipes via pi-super-curl /scurl.
SkillCFreshness: C★ 1Claude Code +9 - 0.491score
ingest-recipes
m4xx101
Tool-ingest discipline for /vibehack-ingest — four modes (CLI/repo/name+desc/inline), validation gate, recipe template.
SkillCFreshness: C★ 1Claude Code +9 - 0.486score
playwright-cli
m4xx101
Browser automation via Playwright CLI.
SkillCFreshness: C★ 1Claude Code +9 - 0.486score
distill-recipes
m4xx101
Pattern extraction for the /vibehack-distill command — situation/action/outcome triples, recipe synthesis.
SkillCFreshness: C★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- May 2, 2026 (5 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/m4xx101/pi-vibehack
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from m4xx101
- 0.496score
curl
m4xx101
HTTP probing recipes via curl.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
operator-recipes
m4xx101
Execution discipline for the pi-vibehack Operator subprocess — evidence capture, replay artifact format, structured-output contract.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
planner-recipes
m4xx101
Reasoning patterns for the pi-vibehack Planner — hypothesis framing, falsifier writing, tree pruning heuristics, recall query crafting.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
reporter-recipes
m4xx101
Writeup style for the pi-vibehack Reporter subprocess — per-leaf PoC format, final report structure, remediation phrasing.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
osint
m4xx101
Specialist for open-source intelligence — domain history, leaked credentials, employee enumeration, infrastructure mapping.
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
web-exploit
m4xx101
Specialist prompt for an Operator subprocess running a web-exploit leaf — auth bypass attempts, RCE chains, deserialization, SSRF.
SkillCFreshness: C★ 1Claude Code +9
Cite this page
browser-verifier skill. Agent Mart. https://agentmart.dev/skills/m4xx101-browser-verifier (last updated 2026-10-24)