opencli-browser

Use when an agent needs to drive a real Chrome window via opencli — inspect a page, fill forms, click through logged-in flows, or extract data ad-hoc. Covers...

Security Scan

The skill's purpose (drive a real Chrome window via opencli) is plausible, but the manifest omits key runtime requirements and the runtime instructions allow broad access to browser page content and network captures — the skill is internally inconsistent and could expose sensitive data if used without care.

Purpose & Capability

The SKILL.md repeatedly instructs the agent to run the opencli CLI and to interact with a running Chrome + extension (e.g., 'opencli doctor', 'opencli browser ...', references to Chrome debug port), but the registry metadata declares no required binaries, no install steps, and no required config paths. A legitimate opencli-based browser driver should at minimum declare opencli (and/or Chrome/Chromium) as a runtime dependency and document any required extension or debug-port config; the absence of those declarations is an incoherence.

Instruction Scope

The instructions themselves stay within the stated purpose (driving a live browser). They explicitly encourage inspecting page state, capturing network requests, running JS eval (read-only), and verifying writes. Those actions necessarily let the agent read page DOM, form contents, and network payloads (potentially including session identifiers or personal data). That is expected for a browser-driving skill, but it means the agent will have access to sensitive user data; the SKILL.md does not provide any additional constraints or warnings about that beyond usage best-practices.

Install Mechanism

This is an instruction-only skill with no install spec or code files. That minimizes disk-write risk. However, because it relies on an external CLI (opencli) and browser/extension setup, the manifest should have declared those runtime requirements.

Credentials

The SKILL.md references environment variables (OPENCLI_WINDOW_FOCUSED, OPENCLI_LIVE) and external browser state, but the declared requires.env is empty. More importantly, the skill's runtime actions (DOM reads, network capture) can expose highly sensitive data — cookies, auth tokens, page content — yet no explicit requirements/consent or warnings are declared in the metadata. The absence of declared binaries/envs and lack of an explicit statement about sensitive data handling is disproportionate to the manifest.

Persistence & Privilege

always:false and default autonomous invocation settings are used (user-invocable, agent may call autonomously). This is the platform default and not itself a problem. Note: because the skill lets the agent control a browser and capture network/page state, autonomous invocation increases risk; exercise caution if you allow the agent to call this skill without manual approval.

What to consider before installing

Before installing or enabling this skill: 1) Confirm you have the opencli CLI and the required Chrome extension/debug-port configuration — the skill's metadata does not declare these but the runtime docs require them. 2) Treat this skill as powerful: it can read page DOM and network traffic (including session tokens or private form data). Avoid running it while logged into sensitive accounts; prefer a dedicated browser profile or test account. 3) Ask the skill author to update the manifest to declare required binaries/env vars and to document how sensitive data is handled. 4) If you must use it, restrict the agent's ability to invoke the skill autonomously or require user confirmation for each run.

The first reader of this CLI is an agent, not a human. Every subcommand returns a structured envelope that tells you exactly what matched, how confident the match is, and what to do if it didn't. Lean on those envelopes — do not guess.

This skill is for driving a live browser to accomplish an agent task. If you are building a reusable adapter under ~/.opencli/clis/<site>/ use opencli-adapter-author instead.

---

Prerequisites

opencli doctor

Until doctor is green, nothing else will work. Typical failures: Chrome not running, extension not installed, debug port blocked by 1Password / other extensions. The doctor output tells you which.

---

Window lifecycle

• opencli browser * commands already keep the automation session alive between calls. The window stays open until you run opencli browser close or the idle timeout expires.
• --focus (or OPENCLI_WINDOW_FOCUSED=1) opens the automation window in the foreground. Use it when you want to watch the page live.
• --live (or OPENCLI_LIVE=1) is mainly for browser-backed adapter commands such as opencli xiaohongshu note .... It keeps the adapter's automation window open after the command returns so you can inspect the final page state.

---

Mental model

1. Selector-first target contract. Every interaction command (click, type, select, get text/value/attributes) takes one <target>, which is either a numeric ref from state/find or a CSS selector. Use --nth <n> to disambiguate multiple CSS matches.
2. Every envelope reports matches_n and match_level. match_level is exact, stable, or reidentified — the CLI already rescued moderate DOM drift for you, but the level tells you how confident to be.
3. Compact output first, full payload on demand. state is a budget-aware snapshot; get html --as json supports --depth/--children-max/--text-max; network returns shape previews and you re-fetch a single body with --detail <key>. If you emit a giant payload you are burning context you did not need to burn.
4. Structured errors are machine-readable. On failure the CLI emits {error: {code, message, hint?, candidates?}}. Branch on code, not on message strings.

---

Critical rules

1. Always inspect before you act. Run state or find first. Never hard-code a ref or selector from memory across sessions — indices are per-snapshot.
2. Prefer numeric ref over CSS once you have it. Numeric refs survive mild DOM shifts because the CLI fingerprints each tagged element. A CSS selector written by hand will break the first time the site re-renders.
3. Read match_level after every write. exact = all good. stable = the element is the same but some soft attrs drifted — your action still applied. reidentified = the original ref was gone and the CLI found a unique replacement; double-check you hit the right element.
4. Use the compound field for form controls. Do not regex-guess a date format, do not state twice to get the full <select> options list. The compound envelope has the format string, full option list up to 50, options_total for overflow, and accept/multiple for <input type=file>.
5. Verify writes that matter. After type <target> <text>, run get value <target>. After select, run get value. Autocomplete widgets, React controlled inputs, and masked fields all silently eat characters. The CLI cannot detect this for you.
6. state → action → state after a page change. Navigations, form submits, and SPA route changes invalidate refs. Take a fresh snapshot. Do not reuse refs from before the transition.
7. Chain with &&. A chained sequence runs in one shell so refs acquired by the first command stay live for the second. Separate shell invocations lose the session context you just set up.
8. eval is read-only. Wrap the JS in an IIFE and return JSON. If you need to change the page, use the structured click / type / select / keys commands instead — they produce structured output and fingerprints, eval does not.
9. Prefer network to screen-scraping. If a page you care about fetches its data from a JSON API, the API is almost always more reliable than scraping the rendered DOM. Capture once, inspect the shape, then --detail <key> the body you need.

---

Target contract (<target> for click / type / select / get text|value|attributes)

<target> ::= <numeric-ref> | <css-selector>

• Numeric ref — the [N] index from state or find. Cheap, resilient to soft DOM drift.
• CSS selector — anything querySelectorAll accepts. Must be unambiguous on write ops, or pair with --nth <n>.

Envelope on success

{ "clicked": true, "target": "3", "matches_n": 1, "match_level": "exact" }

{ "value": "[e-mail]", "matches_n": 1, "match_level": "stable" }

match_level

level	meaning	you should
exact	Fingerprint agreed on tag + strong IDs with at most one soft drift	Proceed.
stable	Tag + strong IDs still agree, soft signals (aria-label, role, text) drifted	Proceed, but if what you typed/clicked matters, re-check with get value or state.
reidentified	Original ref was gone; a unique live element matched the fingerprint and was re-tagged with the old ref	Double-check you hit the right element before chaining more writes.

Structured error codes

Branch on these, not on the human message:

code	meaning
not_found	Numeric ref is no longer in the DOM. Re-state.
stale_ref	Ref exists but the element at that ref changed identity. Re-state.
invalid_selector	CSS was rejected by querySelectorAll. Fix the selector.
selector_not_found	CSS matches 0 elements. Try find with a looser selector.
selector_ambiguous	CSS matches >1 and no --nth. Add --nth or narrow the selector.
selector_nth_out_of_range	--nth beyond match count.
option_not_found	select couldn't find an option matching that label/value. Error envelope includes available: string[] of the real option labels.
not_a_select	select was called on a non-<select> element.

Error envelope always includes error.code and error.message. Target errors (selector_not_found, selector_ambiguous, etc.) often add error.candidates: string[] with suggested selectors. option_not_found adds error.available: string[] instead.

---

Command reference

Inspect

command	purpose
browser state	Snapshot: text tree with [N] refs, scroll hints, hidden-interactive hints, compounds (N): sidecar for date/select/file refs.
browser find --css <sel> [--limit N] [--text-max N]	Run a CSS query and return one entry per match with {nth, ref, tag, role, text, attrs, visible, compound?}. Allocates refs for matches the prior snapshot didn't tag. Cheap alternative to state when you already know the selector.
browser frames	List cross-origin iframe targets. Pass the index to --frame on eval.
browser screenshot [path]	Viewport PNG. No path → base64 to stdout. Prefer state when you just need structure.

Get (read-only)

command	returns
browser get title	plain text
browser get url	plain text
browser get text <target> [--nth N]	{value, matches_n, match_level}
browser get value <target> [--nth N]	{value, matches_n, match_level}
browser get attributes <target> [--nth N]	{value: {attr: val, ...}, matches_n, match_level}
browser get html [--selector <css>] [--as html|json] [--depth N] [--children-max N] [--text-max N] [--max N]	Raw HTML, or structured tree. JSON tree nodes have {tag, attrs, text, children[], compound?}. Truncation reported via truncated: {depth?, children_dropped?, text_truncated?}.

Interact

command	notes
browser click <target> [--nth N]	Returns {clicked, target, matches_n, match_level}.
browser type <target> <text> [--nth N]	Clicks first, then types. Returns {typed, text, target, matches_n, match_level, autocomplete}. autocomplete: true means a combobox/datalist popup appeared after typing — you almost always need keys Enter or a follow-up click on the suggestion to commit the value.
browser select <target> <option> [--nth N]	Matches option by label first, then value. Use compound from find/state to see exactly what labels are available.
browser keys <key>	Enter, Escape, Tab, Control+a, etc. Runs against the focused element.
browser scroll <direction> [--amount px]	up / down. Default amount 500.

Wait

browser wait selector "<css>" [--timeout ms]    # wait until the selector matches
browser wait text "<substring>" [--timeout ms]  # wait until the text appears
browser wait time <seconds>                     # hard sleep, last resort

Default timeout 10000 ms. SPA routes, login redirects, and lazy-loaded lists need wait before state/get.

Extract

• browser eval <js> [--frame N] — Run an expression in the page (or in a cross-origin frame via --frame). Wrap in an IIFE and return JSON. Read-only: no document.forms[0].submit(), no clicks, no navigations. If the result is a string, stdout is the raw string; otherwise it's JSON.
• browser extract [--selector <css>] [--chunk-size N] [--start N] — Markdown extraction of long-form content with a continuation cursor. Returns {url, title, selector, total_chars, chunk_size, start, end, next_start_char, content}. Loop on next_start_char until it is null. Auto-scopes to <main>/<article>/<body> if you don't pass --selector.

Network

browser network                        # shape preview + cache key list
browser network --detail <key>         # full body for one cached entry
browser network --filter "field1,field2"  # keep only entries whose body shape contains ALL fields as path segments
browser network --all                  # include static resources (usually noise)
browser network --raw                  # full bodies inline — large; use sparingly
browser network --ttl <ms>             # cache TTL (default 24h)

List entries look like {key, method, status, url, ct, size, shape, body_truncated?}. Detail envelope is {key, url, method, status, ct, size, shape, body, body_truncated?, body_full_size?, body_truncation_reason}. Cache lives in ~/.opencli/cache/browser-network/ so you can re-inspect without re-triggering the request.

Tabs & session

command	purpose
browser tab list	JSON array of {index, page, url, title, active}. The page string is the tab identity you pass as <targetId> to tab select / tab close, or to --tab <targetId> on any subcommand. (--tab's placeholder is historical — the value is always page.)
browser tab new [url]	Open a new tab. Prints the new page string.
browser tab select [targetId]	Make a tab the default. All subcommands accept --tab <targetId> to target one without changing the default.
browser tab close [targetId]	Close by page.
browser back	History back on the active tab.
browser close	Close the automation window when done.

---

Compound form controls

Every date/time, select, and file input carries a compound field. Use it — do not regex attributes.

Date family

{
  "control": "date",
  "format": "YYYY-MM-DD",
  "current": "2026-04-21",
  "min": "2026-01-01",
  "max": "2026-12-31"
}

control is one of date | time | datetime-local | month | week. format is a concrete template string — type into the field using that exact format, or select by label if the site wraps the native input in a custom widget.

Select

{
  "control": "select",
  "multiple": false,
  "current": "United States",
  "options": [
    { "label": "United States", "value": "us", "selected": true },
    { "label": "Canada", "value": "ca" }
  ],
  "options_total": 137
}

options[] is capped at 50 entries. current is always correct even when the selected option is past the cap — it's computed by scanning every option, not from the truncated list. If options_total > options.length and you need an option that isn't in options[], call browser select <target> "<label>" directly — the CLI matches against the live DOM, not the truncated list.

File

{
  "control": "file",
  "multiple": true,
  "current": ["report.pdf", "cover.png"],
  "accept": "application/pdf,image/*"
}

Do not invent file paths. Upload is done via the normal click flow — respect accept when telling the user what to upload.

Where compounds show up

• browser find --css <sel> entries: inline on each match.
• browser get html --as json tree nodes: inline on matching nodes.
• browser state snapshot: in a compounds (N): sidecar keyed by numeric ref, so you can tell at a glance which [N] entries have rich metadata.

---

Cost guide

Think about payload size per call. Budgets exist for a reason.

command	rough cost	when to use
state	medium (bounded by internal budget)	First call on any page, after every nav, when you need refs.
find --css <sel>	small	You already know the selector — one query, compact entries.
get title / get url	tiny	Sanity checks between steps.
get text/value/attributes	tiny per call	Verifying one specific field.
get html (raw)	can be huge	Avoid on unbounded pages. Always pair with --selector and a budget.
get html --as json --depth 3 --children-max 20	medium	When you need to reason about structure, not a specific field.
screenshot	large	Only when the page is visual (CAPTCHA, charts). Prefer state.
extract	medium per chunk	Long-form reading. Loop via next_start_char.
network (default)	small	First look at APIs.
network --detail <key>	varies	Pull one body.
network --raw	huge	Only after --filter narrowed the candidate set.
eval "JSON.stringify(...)"	controlled	Targeted extraction when none of the above fit.

Rule of thumb: one state per page transition, one find per follow-up query, one get/click/type per action. If your plan involves >10 calls per page you are probably scraping instead of interacting — consider extract or network.

---

Chaining rules

Good — one shell, live session:

opencli browser open "https://news.ycombinator.com" \
  && opencli browser state \
  && opencli browser click 3

Bad — each line is a fresh shell, refs from call 1 are already forgotten when call 2 runs. (Only a problem if you rely on shell-scoped state; browser refs themselves persist in-page, but interleaving unrelated shells invites races.) Prefer && when the steps are meant to be atomic.

Never chain a write and then an immediate state without a wait if the action causes a network round-trip — you will snapshot the pre-response DOM and make bad decisions off stale data.

---

Recipes

Fill a login form

opencli browser open "https://example.com/login"
opencli browser state                          # find [N] for email, password, submit
opencli browser type 4 "[e-mail]"
opencli browser type 5 "hunter2"
opencli browser get value 4                    # verify (autocomplete can eat chars)
opencli browser click 6                        # submit
opencli browser wait selector "[data-testid=account-menu]" --timeout 15000
opencli browser state                          # fresh refs on the logged-in page

Pick from a long dropdown

opencli browser state                          # sidebar shows [12] <select name=country>
opencli browser find --css "select[name=country]"
# the compound.options_total is 137, but compound.current is "" — unselected.
opencli browser select 12 "Uruguay"
opencli browser get value 12                   # { value: "uy", match_level: "exact" }

Scrape a list via network instead of DOM

opencli browser open "https://news.ycombinator.com"
opencli browser network --filter "title,score"
# -> find the /topstories entry, note its key
opencli browser network --detail topstories-a1b2

Read a long article in chunks

opencli browser open "https://blog.example.com/long-post"
opencli browser extract --chunk-size 8000
# -> content + next_start_char: 8000
opencli browser extract --start 8000 --chunk-size 8000
# ...until next_start_char is null

Cross-origin iframe

opencli browser frames
# -> [{"index": 0, "url": "https://checkout.stripe.com/...", ...}]
opencli browser eval "(() => document.querySelector('input[name=cardnumber]')?.value)()" --frame 0

---

Pitfalls

• Do not submit forms via eval "document.forms[0].submit()" — modern sites intercept with JS handlers and silently drop the call. Either click the submit button via its ref, or (if you know the GET URL) just open it directly.
• Do not reuse refs across a page transition. wait for the new state, then re-state. Old refs will either 404 or (worse) reidentify onto a similarly-shaped element on the new page.
• match_level: reidentified is a warning, not an error. The action went through, but if you are chaining 5 more writes that all depend on that being the right element, verify with a get text or get value before continuing.
• Budget-aware commands silently cap. get html --as json with default budgets will return truncated: {...}. If your downstream logic needs the whole subtree, raise --depth / --children-max or tighten the selector.
• autocomplete: true on a type response is not an error. It means a suggestion popup is open and your value isn't committed yet. Typically keys Enter to accept the first suggestion, or click the one you want.
• network --filter is AND-semantics on path segments. --filter "title,score" keeps entries whose body shape contains both title and score as path segments, at any depth. It is not a regex.
• Screenshots are for humans, not for agents. Use state + find unless the page is genuinely visual (captcha, chart). Screenshots burn tokens and rarely add signal an agent can act on.

---

Troubleshooting

symptom	fix
opencli doctor red: "Browser not connected"	Start Chrome with --remote-debugging-port=9222, or rerun the extension install.
attach failed: chrome-extension://...	Disable 1Password / other CDP-hungry extensions temporarily.
selector_not_found right after state	Page mutated. wait selector "..." then retry.
stale_ref across every command	You are reusing refs from a prior page. Re-state.
click succeeds but nothing happens	The element is probably a decorative wrapper stealing clicks from the real target. find --css "..." with a narrower selector and retry on the inner element.
type appears to finish but value is wrong	Autocomplete, masked input, or React controlled re-render. Verify with get value. Add keys Enter or re-type.
Giant get html output	Pass --selector + --as json --depth 3 --children-max 20 --text-max 200.
Network cache seems stale	Bump --ttl down, or let it expire. The cache lives at ~/.opencli/cache/browser-network/.

---

See also

• opencli-adapter-author — turning what you just figured out into a reusable ~/.opencli/clis/<site>/<command>.js.
• opencli-autofix — when an existing adapter breaks, this skill walks you through OPENCLI_DIAGNOSTIC and filing a fix.