Check one page's accessibility: run an audit, read the accessibility tree, and check keyboard navigation and color contrast. Use this for focused accessibility work, and fileforge_audit_action for a wider audit. Doesn't change the page; results can be stored locally. Needs the extension.

Work with the API calls a page makes: discover REST and GraphQL requests, find OpenAPI specs, and export them as curl, Postman or MSW mocks. Reads a running DevTools session, so call start_session on fileforge_devtools_action first or results are empty. Exports replace a file of the same name and can include captured auth headers, so review them before sharing. Needs the extension.

Inspect a web app in an open tab: detect its framework, and read routes, Next.js data, client storage, feature flags and React tree hints. Use this to understand how an app is built. Use fileforge_coding_action for a full bundle. Read-only, but get_client_storage returns raw localStorage and sessionStorage values, which can include tokens. Needs the extension.

Audit a page or site for SEO, content, policy and client-side security issues. Prefer query_audit with a preset; compare_runs shows what changed between stored audits. Use fileforge_a11y_action, fileforge_perf_action or fileforge_responsive_action for one focused check. reveal_client_security_value needs FILEFORGE_ALLOW_SECRET_REVEAL=1. Results can be stored. Needs the extension.

Manage tabs and read browser state: go to a URL, switch, reload and close tabs, move through history, wait for a page, and list open tabs. Use query_context to see what's open, and fileforge_page_interaction to act inside a page. Closing tabs can lose unsaved work, and close_tabs_matching can close many at once, so try dryRun first. Sensitive tab URLs are redacted. Needs the extension.

Get context for fixing code in one call. inspect_page returns a page's framework, selectors, forms, console, network and performance; audit_routes sweeps many routes for errors; bug_context and article_context gather focused context. Use fileforge_qa_testing for pass/fail checks. Opens pages in your session; results can be stored. Needs the extension.

Read DevTools data for a tab, from the moment a session starts: console, JS errors, network requests, API responses, WebSockets and performance. Prefer query_devtools with a preset; the get_* actions are older. Attaches the debugger, so Chrome shows its debugging bar. evaluate runs JavaScript in the page and replay_request re-sends a request; both can change data and have guards. Needs the extension.

Write stored results or SQL query output to a file in formats such as CSV, JSON, Markdown, HAR or ZIP. Use this to hand data to another tool. Use fileforge_result_action or fileforge_sql_action to read data without creating a file. Writes only to FileForge's exports folder, and a filename you reuse replaces that file.

Inspect a Chrome extension: manifest, permissions, storage, service worker, content-script injection and errors. Use this to debug an extension you're building, and fileforge_audit_action for store policy checks. reveal_storage_value needs FILEFORGE_ALLOW_SECRET_REVEAL=1. Needs the extension.

Monitor many tabs at once: network traffic, console errors, JS exceptions, DOM changes and WebSocket frames. Set tripwires such as "status >= 500", then read get_tripwires; they don't notify you. watch_person tags events that mention an identity. Use fileforge_devtools_action for one tab in depth. Attaches the debugger to watched tabs; the incognito actions open and close windows. Needs the extension.

Fetch, parse and convert files, extract text, and decode data URIs, including those in a page. Use this for documents and embedded assets, and fileforge_scrape_action for HTML pages. Fetches public URLs without your browser session; private addresses need FILEFORGE_ALLOWED_FETCH_HOSTS. Local paths need the output folders of ~/.fileforge or FILEFORGE_ALLOWED_READ_PATHS. Can write files locally.

Remember pages over time: save a page, find earlier visits, diff it against a previous visit, and search saved pages or pull evidence for a question. Use this to see what changed on a page. Use fileforge_result_action to read tool results and fileforge_search_action to search history. Data stays in ~/.fileforge. save_page and diff_page need the extension.

Keep local notes: create, list and search them. Use this to save findings across sessions. Use fileforge_memory_action to save page content and fileforge_report_action for shareable evidence. Notes stay in ~/.fileforge, and there's no update or delete action.

Read and act inside one open page: get text, snapshots and forms, click, fill, scroll, assert, and debug selectors. click and fill act on the real signed-in page, so they can submit or change data. hide_elements lasts until restore_hidden_elements. Use fileforge_qa_testing for repeatable checks across pages and fileforge_scrape_action for structured extraction. Needs the extension.

Run a performance audit on a URL across viewports and return Core Web Vitals. Use this for a performance check of one page. Use fileforge_devtools_action for raw session data and fileforge_audit_action for SEO, content and security. Loads the URL in its own tab, closes it afterward, and stores the result by default. It doesn't change the site. Needs the extension.

Run repeatable pass/fail checks: smoke tests, page boot, routes, console, network, forms and element watches. Use fileforge_page_interaction for individual steps and fileforge_coding_action while fixing code. URL checks load pages in your session. check_console_clean and check_network_clean read a running DevTools session and return SESSION_NOT_STARTED without one. Needs the extension.

Generate evidence bundles from the active tab: a bug report, PR evidence with screenshots and QA results, or a full debug bundle. Use this when you need something to share or attach. Use fileforge_coding_action for context while working. Bundles are saved locally and can contain page data, so review them before sharing. Needs the extension.

Find layout bugs across screen widths: scan a URL, and locate overflow and overlapping elements. capture_matrix takes a screenshot at each viewport. Use fileforge_visual_action to compare screenshots against baselines. Loads the page in your session and can store results; it doesn't change the site. Needs the extension.

Read results other tools stored in ~/.fileforge: fetch one by ref, list them, search them, or slice a large one. Use this when a tool returns a ref instead of full data. Use fileforge_sql_action to query across results and fileforge_export_action to write one to a file. Works without the extension.

Check FileForge itself: whether the server and extension are connected, their versions, which tools are available, and recent requests. Use status when other tools report a bridge or connection error. It doesn't touch page content.

Extract structured content from pages, tabs or a whole site: text, links, tables, metadata, assets and design tokens. Prefer query_scrape; the others are older single-purpose actions. It can't click or submit; use fileforge_page_interaction for that. crawl_site opens real tabs in your signed-in session, up to 50 pages by default. Query stored links and tables with fileforge_sql_action. Needs the extension.

Search local data by text: saved pages, stored results, notes, browser history and bookmarks. Use all to search everything, or narrow it to one source. Use fileforge_memory_action to compare page visits and fileforge_scrape_action to read a live page. History and bookmarks need the extension.

Query FileForge's stored results with SELECT-only SQL. Use list_tables and describe_table to see the schema, query to read rows, and export_query to write rows to a file. Covers saved results and scraped pages, links and tables, not the live browser. Use fileforge_result_action to fetch a single known ref. Works without the extension.

Run multi-step work two ways. Tasks (start, add, status, monitor_page) track one job. Routines (create, validate, dry_run, run, plan_from_prompt) are saved and can be scheduled, but cron only fires while this server runs. For a single step, call the tool directly. Runs act in your browser, pausing only at steps that require approval. delete removes a task; routines can only be disabled.

Capture and compare how pages look: screenshots of a page, element or region, baselines, visual diffs across routes and viewports, and computed CSS, colors and fonts. Use fileforge_responsive_action to find layout bugs. Files and baselines are saved locally. capture_screenshot needs the tab to be visible. Needs the extension.

Record browser actions as a reusable recipe, then list, replay, export or delete recipes. Use this to repeat a sequence you did by hand. Use fileforge_task_action for scheduled or multi-step routines. Replaying really clicks and types in your signed-in browser, and delete_recipe is permanent. Needs the extension.