libretto-readonly
libretto-readonly lets you examine and diagnose the current state of any live browser session without altering the page or application. Built as part of Saffron Health's AI automation toolkit, it provides reliable inspection capabilities for debugging and understanding browser behavior in real time.
libretto-readonly lets you examine and diagnose the current state of any live browser session without altering the page or application. Built as part of Saffron Health's AI automation toolkit, libretto-readonly provides reliable inspection capabilities for debugging and understanding browser behavior in real time. It supports focused inspection queries like checking visibility, text content, and element counts, making it ideal for diagnosing issues without side effects.
AI-generated summary based on this skill's SKILL.md
Decision gist · record as of 2026-07-27
libretto-readonly lets you examine and diagnose the current state of any live browser session without altering the page or application. Built as part of Saffron Health's AI automation toolkit, libretto-readonly provides reliable inspection capabilities for debugging and understanding browser behavior in real time. It supports focused inspection queries like checking visibility, text content, and element counts, making it ideal for diagnosing issues without side effects.
Use it when
- Yes.
- libretto-readonly enables you to debug failed browser automation jobs by connecting to an existing browser session in read-only mode.
Verify before relying
Read SKILL.md below before installing (1 file). Open directory: indexed for reading, not audited.
Install
saffron-health/libretto/libretto-readonly · repository language: TypeScript
Open directory. Skills are indexed for reading, not audited. Review a skill's body before installing it.
Frequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
What is libretto-readonly and what can it do?
libretto-readonly lets you examine and diagnose the current state of any live browser session without altering the page or application. Built as part of Saffron Health's AI automation toolkit, libretto-readonly provides reliable inspection capabilities for debugging and understanding browser behavior in real time. It supports focused inspection queries like checking visibility, text content, and element counts, making it ideal for diagnosing issues without side effects.
Can I diagnose live browser state without clicking or modifying the page?
Yes. libretto-readonly is specifically designed to diagnose and inspect live browser state without making any changes to the page or application. You can view page content, DOM structure, and HTTP responses safely in read-only mode. This non-destructive browser inspection approach preserves the current state of your browser session, allowing you to understand what's happening without introducing unintended side effects.
How does libretto-readonly help debug failed browser automation jobs?
libretto-readonly enables you to debug failed browser automation jobs by connecting to an existing browser session in read-only mode. Once connected, you can view page content, DOM, and HTTP responses to diagnose what went wrong. This approach lets you inspect the exact state the browser was in when a job failed, making root-cause analysis faster and more reliable without risking further state changes.
What inspection queries can libretto-readonly perform?
libretto-readonly supports focused inspection queries including checking element visibility, extracting text content, counting elements, and examining DOM structure. These queries allow you to gather diagnostic information about your page state in a targeted way. Combined with access to console output and HTML snapshots, libretto-readonly gives you comprehensive visibility into browser behavior for effective debugging.
How do I connect to an existing browser session using libretto-readonly?
libretto-readonly lets you connect to an existing browser session in read-only mode to preserve its current state. This connection approach is designed for inspection workflows where you want to examine a live session without triggering any interactions or mutations. After diagnosis is complete, you can close the inspection session to avoid resource waste.
Is libretto-readonly free to use?
Yes, libretto-readonly is released under the MIT license, making it free and open source. You can use it for both personal and commercial projects without licensing restrictions. The MIT license provides full transparency and allows you to modify and distribute the tool as needed for your browser inspection and debugging workflows.
SKILL.md
Rendered from the published skill. Quoted content, verbatim.
How Libretto Read-Only Works
- Use this skill when the browser session must stay strictly read-only.
- Libretto stores read-only vs write-access on the session itself.
- The primary inspection tools are
snapshotandreadonly-exec. readonly-execreuses Libretto's normal execution pipeline, but it only exposes read-only helpers and denies mutating Playwright methods.- Only a user can change the session mode for an existing session. Never change a session's mode on your own — the user must change it themselves manually.
Working Rules
- Announce which session you are using and what page you are inspecting.
- Do not use
exec,run, or any direct Playwright action that could change browser or application state. - Do not click, type, submit forms, navigate, upload files, dispatch DOM events, or send non-GET requests.
- Prefer
snapshotfirst when the visible page state is unclear. - Use
readonly-execfor focused inspection: titles,
(truncated - see the full file via the links below)
File tree — 1 file
.agents/skills/libretto-readonly/SKILL.md
Let your AI agent find skills like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 56,283 agent skills by what they can do, searchable in plain language.
wish › “Diagnose and inspect live browser state without making any changes to the page or application”
Give your agent the search over MCP, or paste the wish link into any chat. No install? Search from any chat →
Related skills
dev-browser equips you with AI-driven tools to construct robust browser automation workflows. Streamline web navigation, data extraction, and interaction tasks with a toolkit designed for reliability and consistency across complex automation scenarios.
Control Electron-based desktop applications programmatically by connecting through Chrome DevTools Protocol. This skill enables automation of popular tools like Slack, Discord, and VS Code, letting you interact with their UI elements and workflows as if you were using them directly. Built on reliable browser automation foundations for consistent, repeatable interactions.
This skill teaches you battle-tested approaches to structuring and scaling Playwright test automation projects. Learn how to write maintainable test code, organize test suites effectively, and apply design patterns that reduce technical debt and improve test reliability across your automation framework.
Control browser sessions over HTTP to test pages interactively, run reusable automation flows, and validate multi-user scenarios with different auth profiles. Execute Playwright code against live or local pages, capture screenshots, and persist authentication state across sessions.
This skill enables you to write and run Playwright test scripts within Claude Code, automating verification of web application UI behavior and user interactions. Deploy comprehensive browser-based testing workflows without leaving your coding environment.
This skill provides direct access to the Retool PostgreSQL database, letting you retrieve moderation notes, user notes, and other data managed through Retool dashboards. Queries run read-only by default for safety, with write operations available only when explicitly enabled and permitted. Use it to explore table schemas, look up user-specific notes, or investigate Retool-managed records.