Step-by-step guide
HermodLens
Scope PDF accessibility remediation from documents linked on selected website pages.
For small public-sector web teams or accessibility agencies scoping document remediation.
See it in use
Inspect a document scope and estimate
A screenshot walkthrough with fictional practice data. Select a step to explore.
1. Start with
Inspect a document scope and estimate
The free pack contains a 17-document fictional report and register.
Sample files and contents
Free illustrated guide, fictional two-host practice site, and actual HTML/CSV/JSON outputs.
Practice files and guides are free. The complete app is included only where the product page says so.
2. In the app
Try it in three steps
- Open the saved offline HTML report.
- Review the page with no links and nine records outside the estimate.
- Compare the register with the estimate assumptions.
Show every click and setting
- Open examples/sample-exports/hollowmere-report.html locally.
- Read the page that returned no document links and the nine document records outside the estimate.
- Compare the register with the estimate assumptions. Use the matching JSON later in the installed extension.
3. The result
Inspect the result
The handoff lists 17 document records and outstanding review. New scans require the separate extension.
What to check in the practice result
A concrete handoff with 17 document records, confirmed example rates and outstanding review. New scans require the separately sold extension.
This fictional example is shown in the working interface and in the supplied result. Review the guide before using your own files.
See the workflow at a glance
Choose a short explanation, then follow the worked example below. Check the requirements and compatibility before using your own files.
Scope the document work without hiding the unknowns.
A remediation estimate needs a document register and visible assumptions. HermodLens collects linked PDFs and Office files from selected pages, inspects PDFs locally and prepares a register with a page-based estimate using your rates. The fictional Hollowmere exercise keeps skipped or inaccessible documents visible as unassessed. A monetary range appears only after you confirm the assumptions. This is triage, not accessibility certification or a legal opinion. The interactive exercise requires the purchased unpacked Chrome extension; browser policy may block installation.
View the illustrated card
Check the fit before you buy
Inspect the saved scope, confirm installation is allowed, then decide whether this workflow suits your work.
Read the actual output
The free report shows what was found, what remains outside the estimate and who needs to review it.
Run your own scope
The paid download includes the unpacked extension, source, illustrated guide, practice site and saved outputs.
Confirm the limits
Check desktop browser policy and plan for human review. This tool does not certify accessibility.
Before you start
- Desktop Chrome with Developer mode and unpacked extensions allowed. Version 1.0.1 tested in Chrome for Testing 151.0.7922.34 on Linux x64.
- Page URLs, document URLs or local files you are authorised to assess. Downloads are anonymous, without browser cookies.
- Optional Node.js 22 or later to serve the fictional practice website. No Node.js is needed for the installed extension.
Your first result
Install and keep the folder
Extract HermodLens-1.0.1.zip into a folder you will keep. Read START-HERE.html. On chrome://extensions, turn on Developer mode, choose Load unpacked and select its extension/ folder. This is not a Chrome Web Store install. Open HermodLens from the toolbar or with Alt+Shift+H.
Practise on the fictional site
From the extracted bundle, run node tools/serve.mjs and paste the six printed URLs into Pages to scan. Allow the first local host, then the second document host when asked. The free guide also includes this fictional practice site and saved outputs.
Check coverage before estimating
Scan the intended pages, pasted document URLs or local files. Inspect Scope coverage first: follow up failed pages, pages with no links and every document outside the estimate. For a JavaScript-built list, open the page and use Collect document links from this tab. Direct-file projects clearly say no pages were scanned.
Review the register and assumptions
Open a document’s Details or choose Review in register from coverage. Add decisions, service-critical flags and notes. Enter your own low/high rates, currency and included decisions, then confirm them to show the estimate. Unanalysed PDFs, Office files and other excluded records stay visible for follow-up.
Export a usable handoff
Download CSV, the self-contained HTML report, and Project JSON. The HTML keeps page coverage and excluded-document follow-up even if a scan finds no documents. On narrow windows it uses labelled document cards; summary and estimate tables have keyboard-focusable scroll regions. Print from the browser if you need PDF.
Back up before updating
Export JSON before loading a new version or removing an extension. A new unpacked folder may receive a different extension ID and separate storage. Import the saved project into the new installation and check it before removing the old one. Exported files remain yours to keep.
A worked example
Inspect a 17-document scope before buying
Input: Free illustrated guide, fictional two-host practice site, and actual HTML/CSV/JSON outputs.
- Open examples/sample-exports/hollowmere-report.html locally.
- Read the page that returned no document links and the nine document records outside the estimate.
- Compare the register with the estimate assumptions. Use the matching JSON later in the installed extension.
Expected result: A concrete handoff with 17 document records, confirmed example rates and outstanding review. New scans require the separately sold extension.
More application screens
Know the scope before you quote
Real Chrome 151/Linux captures of version 1.0.1, using the included fictional website and its saved outputs.
1. See what the estimate leaves out
2. Turn each finding into a decision
3. Hand over a report people can use
Troubleshooting
Load unpacked is unavailable
Your organisation may block unpacked extensions. Ask your IT team before buying; managed-browser policy was not verified.
A page has no document links
Check the page yourself. For a JavaScript-built list, collect links from the live tab. No links found is not proof that a site has no documents.
Files need a login or are protected
Downloads do not use browser cookies. Review Cannot assess records; obtain authorised local copies or an unlocked file from its owner.
No money totals appear
Review and confirm your own rates, currency and included decisions. Editing assumptions clears confirmation. Unconfirmed exports omit money totals.
The old project disappeared after updating
Import the JSON backup into the new extension installation. Different unpacked folders can have different IDs and storage. Export before removing the old installation.
Compatibility & limits
- Verified on Chrome for Testing 151.0.7922.34/Linux x64. Windows, macOS, Edge, other Chrome versions and managed-browser policy remain unverified. Narrow-window checks are not mobile installation support.
- Machine triage only: no accessibility certification, legal conclusion or promise that every issue is found. A person must review each document.
- Only supplied pages, live-tab links, pasted URLs and local files are in scope. This is not a recursive whole-site crawler. Downloads are anonymous; protected or inaccessible items need follow-up.
- Only analysed PDFs included by your decisions contribute to estimates. Office documents, inaccessible files, duplicates and exclusions remain listed. Rates are your assumptions, not a market quote.
- Fictional native verification reached 17 document records across the tested paths. Customer sites, maximum 300-page/2000-document capacity and other environments were not verified.
Need a hand?
Tell us the product, host application and version, what you tried, and the exact error. Contact support with a fictional example; keep private customer files and passwords out of your message.