Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Browser Control Skill · wecode-ai | DecimalAI

researchby wecode-ai

Browser Control Skill

wecode-ai/browser

Source wecode-ai/wegent ↗License Apache-2.0

Complete real user web tasks end-to-end via browser-tool, navigate, interact, wait for page state, extract results, and provide evidence when needed.

SkillScore57Maintained

Automatic

QA & Test Engineering pack

Last verified 8/18/2026

What it changes

Passed 8 of 14 test cases with the skill, and 7 of 14 without it.

Without skill48%

With skill57%

Checks it passed that the model alone missed

Issues an act request with kind set to close.
Runs a snapshot command with interactive set to true.
Passes ref set to 'e15' inside the evaluate request payload.

and 2 more in the full report

5 fixed8 passed either way7 still failing3 broke

Pass-rate lift: +9 pp
Tokens: +32%
Agent turns: 0%
Response time: -36%

Model gemini-3.6-flashMeasured 8/18/2026v113 of 23 cases passed with skill · 14 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/wecode-ai-browser/SKILL.md https://app.decimal.ai/s/wecode-ai-browser/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add wecode-ai-browser, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/wecode-ai-browser/ and pins it in .decimal/skills.lock

decimalai skills add wecode-ai-browser

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/wecode-ai-browser/SKILL.md https://app.decimal.ai/s/wecode-ai-browser/SKILL.md

Raw file: /s/wecode-ai-browser/SKILL.md ↗ · machine-readable summary at /s/wecode-ai-browser.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 5

Prompt

Show the exact CLI command line execution string to navigate to https://news.ycombinator.com using the browser tool executable.

What was checked, without and with the skill
Checked	Without skill	With skill
Formats the command using the browser-tool CLI binary passing a JSON payload string.	Missed	Passed

case-21 · gemini-3.6-flash · 8/18/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md90 linessingle file

Browser Control Skill

Goal

Finish the user’s real task reliably.
Prioritize successful completion and correct results over aggressive call minimization.

Operating Rules

Start with the intended action directly (navigate/open/act/evaluate). Do not run status as a pre-check.
Use snapshot only when refs are required for interaction (click/type/select/drag/scrollIntoView).
Prefer evaluate for extraction. Return structured data in one comprehensive call when possible.
Use condition waits by default (loadState/url → selector/text/textGone → fn). Avoid timeMs unless explicitly needed.
Before clicking potentially off-screen elements, run act.scrollIntoView on the ref first.
Keep context stable: once targetId is known, pass it in follow-up calls when supported.
Avoid blind loops: every extra call must have a clear purpose.

Reliability and Recovery

If Ref not found, do not reuse stale refs. Take one fresh snapshot, retry once, then stop if still failing.
For repeated failures with the same cause, stop and explain the blocker clearly instead of retrying endlessly.
Connection recovery is built into the tool. Allow auto-recovery once; if still disconnected, instruct user to install/connect extension.

Screenshot Policy

Default: no screenshot.
Use screenshots only when user asks, or when visual proof is required.
Prefer element screenshots (ref or element) over full-page screenshots.
Use full-page screenshots only for page-level evidence.

Recommended Flow

Direct action first (navigate/open or immediate act/evaluate).
If interaction needs refs, run snapshot (interactive: true preferred).
Wait for readiness using act.wait with explicit conditions.
Interact (scrollIntoView → click/type/select/drag as needed).
Extract/verify with evaluate (preferred) or snapshot.
Provide screenshot evidence only when necessary.

Connection Handling

Track record

Benchmark

8 / 14

test cases passed, vs 7 / 14 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

wecode-ai / wegent

backend/init_data/skills/browser/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/3/2026Version v1

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Browser Control Skill

Goal

Operating Rules

Reliability and Recovery

Screenshot Policy

Recommended Flow

Connection Handling

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills