Agent Browser
Use this skill for shell-driven browser automation with the local agent-browser CLI. It is most useful for website-access audits, form/login checks, screenshots, and extracting visible page text when curl is not enough.
Prefer the Codex Browser plugin when it is available and the target can be handled in-app. Use this skill when you need repeatable CLI automation, session persistence, downloads, or direct page text/screenshot artifacts from shell.
Core Loop
- Normalize the target URL.
- Open the page.
- Wait for load or a bounded timeout.
- Capture status evidence: URL, title, screenshot, and visible text.
- Classify the outcome.
- Retry with a fallback method only when the first failure is inconclusive.
Essential Commands
agent-browser open "https://example.com"
agent-browser wait --load networkidle
agent-browser get url
agent-browser get title
agent-browser get text body
agent-browser screenshot --full
agent-browser snapshot -i
For interactive pages:
agent-browser snapshot -i
agent-browser click @e1
agent-browser fill @e2 "value"
agent-browser press Enter
agent-browser wait --load networkidle
Website Access Classification
Use these statuses for platform audits:
OK_PUBLIC: page loads and meaningful public content is extractable.OK_PARTIAL: page loads but key data is hidden, paginated, script-heavy, or incomplete.LOGIN_REQUIRED: login wall blocks the needed information.BOT_BLOCKED: CAPTCHA, Cloudflare challenge, 403 anti-bot, or similar block.GEO_BLOCKED: explicit country/region restriction.TIMEOUT: page did not load within bounded retries.DOWN: DNS/TLS/connection failure or 5xx confirmed after retry.UNKNOWN: conflicting evidence; manual review needed.
Safety And Credentials
- Never print passwords, cookies, tokens, or private session state.
- If credentials are provided, prefer encrypted
agent-browser auth or session state files outside project deliverables.
- Do not bypass access controls, paywalls, CAPTCHAs, or account restrictions.
- Public extraction is allowed; authenticated extraction requires valid credentials and a legitimate use case.
When to Use vs Alternatives
browser: use first for normal Codex in-app website inspection and visible browser work.agent-browser: use for repeatable shell-based access tests, screenshots, downloads, or batch page extraction.apify-ultimate-scraper: use when platform-specific scraping actors exist and credentials/API tokens are available.dogfood: use for systematic UX/bug exploration of one web app, not for a broad list of affiliate platforms.
1---2name: agent-browser3description: Control websites through the local agent-browser CLI for navigation, screenshots, form interaction, login flows, text extraction, and website-access testing. Use when auditing whether websites are reachable or extractable, when the Codex in-app Browser plugin is unavailable/insufficient, or when a repeatable shell-driven browser workflow is needed. ANTI-PATTERN: Do not use for pure HTTP checks that curl can answer, source-code review, or restricted authenticated content without credentials.4---56# Agent Browser78Use this skill for shell-driven browser automation with the local `agent-browser` CLI. It is most useful for website-access audits, form/login checks, screenshots, and extracting visible page text when `curl` is not enough.910Prefer the Codex Browser plugin when it is available and the target can be handled in-app. Use this skill when you need repeatable CLI automation, session persistence, downloads, or direct page text/screenshot artifacts from shell.1112## Core Loop13141. Normalize the target URL.152. Open the page.163. Wait for load or a bounded timeout.174. Capture status evidence: URL, title, screenshot, and visible text.185. Classify the outcome.196. Retry with a fallback method only when the first failure is inconclusive.2021## Essential Commands2223```bash24agent-browser open "https://example.com"25agent-browser wait --load networkidle26agent-browser get url27agent-browser get title28agent-browser get text body29agent-browser screenshot --full30agent-browser snapshot -i31```3233For interactive pages:3435```bash36agent-browser snapshot -i37agent-browser click @e138agent-browser fill @e2 "value"39agent-browser press Enter40agent-browser wait --load networkidle41```4243## Website Access Classification4445Use these statuses for platform audits:4647- `OK_PUBLIC`: page loads and meaningful public content is extractable.48- `OK_PARTIAL`: page loads but key data is hidden, paginated, script-heavy, or incomplete.49- `LOGIN_REQUIRED`: login wall blocks the needed information.50- `BOT_BLOCKED`: CAPTCHA, Cloudflare challenge, 403 anti-bot, or similar block.51- `GEO_BLOCKED`: explicit country/region restriction.52- `TIMEOUT`: page did not load within bounded retries.53- `DOWN`: DNS/TLS/connection failure or 5xx confirmed after retry.54- `UNKNOWN`: conflicting evidence; manual review needed.5556## Safety And Credentials5758- Never print passwords, cookies, tokens, or private session state.59- If credentials are provided, prefer encrypted `agent-browser auth` or session state files outside project deliverables.60- Do not bypass access controls, paywalls, CAPTCHAs, or account restrictions.61- Public extraction is allowed; authenticated extraction requires valid credentials and a legitimate use case.6263## When to Use vs Alternatives6465- `browser`: use first for normal Codex in-app website inspection and visible browser work.66- `agent-browser`: use for repeatable shell-based access tests, screenshots, downloads, or batch page extraction.67- `apify-ultimate-scraper`: use when platform-specific scraping actors exist and credentials/API tokens are available.68- `dogfood`: use for systematic UX/bug exploration of one web app, not for a broad list of affiliate platforms.
cryptopafi