Browser Agent Workflow Automation Skill

Overview

This skill provides a practical framework for building browser-driven agent workflows that survive real-world UI variability. It prioritizes deterministic automation, explicit checkpoints, and safe failure handling instead of brittle click-recording scripts.

Compatibility

Native

• Claude Code / Claude: native skill usage via SKILL.md.
• Codex/OpenAI workflows: compatible with Agent Skills-style SKILL.md content as reusable workflow instructions.

Manual Adaptation

• Gemini CLI: native skill usage via .gemini/skills/<skill-name>/SKILL.md or .agents/skills/<skill-name>/SKILL.md where supported.
• Cursor: use the generated .cursor/rules/*.mdc adapter for project rules.
• OpenClaw and similar agents: use the same skill content as a reusable prompt/workflow file when native skill import is unavailable.

Prerequisites

• Clearly defined desired outcome and termination condition
• Stable environment (staging/prod) with acceptable automation policy
• Credential/session strategy for authenticated steps

Workflow Blueprint

• Define start state and target success state
• Use resilient selector hierarchy (semantic selectors first)
• Add validation checkpoints between major steps
• Use bounded retries with explicit failure reasons

How to Use This Skill

Prompt Pattern

Apply browser agent workflow automation skill.
Deliver:
1) deterministic step plan,
2) selector map,
3) retry/timeout policy,
4) failure recovery playbook.

Execution Flow

1. Map the workflow into atomic actions and checks.
2. Choose reliable selectors and alternates.
3. Add guardrails for destructive actions (confirmations/human gate).
4. Instrument logs/screenshots for every failed checkpoint.
5. Run repeatedly to validate stability before shipping.

Troubleshooting

Issue: Flaky failures on dynamic pages
Fix: replace brittle selectors with role/label/text anchors and add readiness checks.

Issue: Infinite retries hide root cause
Fix: cap retries and emit structured failure reasons after each attempt.

Issue: Automation performs unsafe actions
Fix: require explicit confirmation gates for irreversible operations.

Knowledge Freshness

Treat tooling details as time-sensitive. Re-validate APIs, limits, pricing, auth models, and deployment flags immediately before implementation. If docs conflict with prior memory, follow current official docs and release notes.

Retrieval Sources

• https://playwright.dev/docs/best-practices

Output Contract

1. Return a concrete plan with implementation order.
2. Provide production-ready commands/config/code snippets (not placeholders).
3. Include explicit assumptions and unresolved risks.
4. Include a verification checklist with pass/fail criteria.

Quality Gates

• All commands are copy/paste ready.
• Security-sensitive steps call out secret handling and least privilege.
• Version-sensitive guidance cites current docs used.
• Rollback path is included for risky changes.
• Final output includes quick validation commands/tests.