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/browseragentprotocol/bap/bap-browsernpx skills add browseragentprotocol/bap --skill bap-browsergit clone --depth 1 https://github.com/browseragentprotocol/bapWhat 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 | $0.00056 | $0.02265 |
| Opus 5 | $0.00028 | $0.01132 |
| Sonnet 5 | $0.00011 | $0.00453 |
| Haiku 4.5 | $0.00006 | $0.00227 |
Grade A, and why
bap-browser 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 2d 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 — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BAP Browser CLI
AI-first browser automation. Like playwright-cli but with composite actions, semantic selectors, self-healing selectors, and action caching built in.
Quick Start
bap demo # Guided walkthrough for first-time users
bap goto https://example.com --observe # Navigate + observe in 1 fused call
bap click role:button:"Get Started" # Semantic selector
bap close
Inside this repo, use pnpm exec bap instead of bare bap. Prefer bap goto for navigation; use bap open only for explicit browser lifecycle control.
Smart Defaults
Session persistence — Browser pages persist across CLI invocations. Disconnect parks context; reconnect restores it. Named sessions: bap -s=checkout goto /cart. Auto-expire after 5 minutes.
Self-healing selectors — When a selector fails, BAP tries fallback identity signals (testId, ariaLabel+role, id, name) before erroring. No flags needed.
Action caching — Selector resolutions cached to ~/.bap/cache/actions/ (24h TTL). Repeat actions skip re-resolution.
Usage
Composite actions — multiple steps in ONE command:
bap act fill:role:textbox:"Email"="[email protected]" \
fill:role:textbox:"Password"="secret" \
click:role:button:"Sign in"
Step syntax: action:selector=value or action:selector.
Fused operations — combine server calls into one, cutting roundtrips:
bap goto https://example.com --observe # Navigate + observe (1 call, not 2)
bap act click:role:button:"Submit" --observe # Act + post-observe (1 call)
bap observe --diff # Incremental: only changes
bap observe --tier=minimal # Minimal response (refs + names only)
Always prefer fused calls. --observe saves a roundtrip. --diff avoids re-scanning unchanged elements. --tier=minimal reduces response size.
Common patterns:
bap act click:text:"Accept" goto:https://example.com/app # Dismiss + navigate
bap act fill:role:searchbox:"Search"="query" press:Enter # Search
bap goto https://app.example.com/login --observe # Login flow (2 calls)
bap act fill:label:"Email"="[email protected]" fill:label:"Password"="s" \
click:role:button:"Sign in" --observe
What ships with it
3 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.
- 2d ago First seen · 214 lines · 56 tokens per session scan A c8fd70e74fbc
bap-browser is a skill published in the GitHub repository browseragentprotocol/bap (5 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 56 tokens to every session and 2,265 once invoked, about $0.0003 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.
moli-webfetch
Fetch, inspect, crawl, and capture live, JavaScript-rendered websites with Moli. Use when Codex needs current web content, web research, fact lookup, link following, a bounded crawl, client-rendered or response-gated content, network diagnostics, or a standalone HTML, Markdown, JSON, semantic-tree, viewport or…
moli-cdp-server
Start Moli's CDP server and connect Playwright, Puppeteer, or raw CDP clients. Use to run a headless-browser CDP endpoint, replace a Chromium process, attach over CDP, enable real layout and screenshot surfaces, or diagnose CDP discovery, connection, and target startup—even when Moli is not named.
playwright-rs-usage
Procedural reference for using playwright-rs in Rust browser-automation code — object model (Browser/Context/Page/Locator), the locator!() macro, builder pattern for options, auto-wait semantics, adding the crate and installing its browsers, and how to capture / inspect traces for failure diagnosis. Use when writing…
opensteer
Direct browser control via CDP. Use when the user wants to automate, inspect, scrape, test, or interact with web pages.
apitap
ApiTap gives AI agents cheap access to web data through three layers.