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 skills add Akakaui/visual-browser-agent --skill workflow-observationgit clone --depth 1 https://github.com/Akakaui/visual-browser-agentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/akakaui/visual-browser-agent/workflow-observation)<a href="https://agentmods.dev/skills/akakaui/visual-browser-agent/workflow-observation"><img src="https://agentmods.dev/badge/skills/akakaui/visual-browser-agent/workflow-observation.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00037 | $0.00759 |
| Opus 5 | $0.00018 | $0.00380 |
| Sonnet 5 | $0.00007 | $0.00152 |
| Haiku 4.5 | $0.00004 | $0.00076 |
Grade A, and why
workflow-observation 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 6d 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workflow Observation Skill
When to Use
Use this skill to understand how a workflow behaves end to end:
- Onboarding, signup, search, cart, or checkout flows
- Multi-step forms and wizards
- Documenting a flow before redesign or test automation
- Finding confusing or failure-prone stages in a UX walkthrough
Instructions
Preparation
- Confirm the environment with the user: production data vs sandbox/test account
- State the contract up front: observe and map, never transact
- Break the workflow into named stages (e.g., landing -> search -> detail -> cart -> checkout)
Stage Recording
For each stage:
record_interaction(action: "start")before the first step of the stage- Perform minimal steps (
click,fill) needed to progress - Keep clips short: stop after the stage completes (max ~15 seconds)
capture_screenshotat every decision point and state change
Step Extraction
From clips and screenshots, produce a numbered step map:
- Each entry: action taken -> expected result -> evidence ID (clip/screenshot)
- Mark branches, optional paths, back-navigation, and dead ends
- Note wait states, spinners, and redirects between steps
Friction Identification
Flag stages that are:
- Confusing: ambiguous labels, hidden primary actions, unexpected navigation, jargon
- Failure-prone: validation appearing only at the end, destructive actions adjacent to safe ones, no undo, vague error messages, lost input on back-navigation
Each flagged stage cites the clip timestamp or screenshot that shows the problem.
Hard Boundaries
- NEVER submit final forms (order placement, payment, account changes, publishing, deletion) without explicit user approval for that exact action
- Fill only clearly test-safe fields; use obvious placeholder values ("Test User", "[email protected]")
- If a stage cannot be observed safely, record it as UNOBSERVED rather than forcing through it
Safety
- Observation mode only: no purchases, submissions, or state-mutating confirmations
- Prefer sandbox/staging environments; ask when unsure which environment you are in
- Do not enter real personal data, payment details, or credentials into observed flows
- Stop immediately if an action would create legal or financial obligations
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.
- 6d ago First seen · 85 lines · 37 tokens per session scan A 18c7a77fc143
workflow-observation is a skill published in the GitHub repository Akakaui/visual-browser-agent (0 stars, last pushed 12d ago), licensed MIT. It adds 37 tokens to every session and 759 once invoked, about $0.0002 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
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
interactive-login
How to complete browser/interactive logins (aws / gh / glab / gcloud). The platform backgrounds the login poller so it survives the human's browser round-trip — and when that does NOT work.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
azure-messaging-webpubsub-java
Build real-time web applications with Azure Web PubSub SDK for Java. Use when implementing WebSocket-based messaging, live updates, chat applications, or server-to-client push notifications.
google-safe-browsing
Prevent and fix Google Safe Browsing "Dangerous site" flags. Use when launching a public web app, buying/picking a domain, building a login or signup page, or when any site shows a red "Dangerous site" / "Deceptive site" warning in Chrome, Brave, Safari, Firefox, or Edge. Triggers on "dangerous site", "deceptive…