The X icon looked like a dismiss/close button (temporary). Changed to
a small 'IGNORE' text link that clearly communicates the action is
permanent — the value will never be flagged again.
https://claude.ai/code/session_01SWSwDfMVij53bCTNSCLMwn
Proper noun heuristic:
- Changed from matching any single capitalized word to requiring
TWO OR MORE consecutive capitalized words (e.g. "Acme Corp")
- Single capitalized words at sentence starts were causing massive
false positives — every sentence starts with a capital letter
- Minimum 5 characters total and 2 proper words required
Ignore button:
- Each PPI warning item now has an X (ignore) button alongside
the + (add mapping) button
- Ignored values are persisted to storage (ss_ignored_ppi) so
they stay dismissed across page reloads
- Ignored values are skipped in both pattern detection and
proper noun detection
- Clicking ignore removes the item from the warning and re-scans
https://claude.ai/code/session_01SWSwDfMVij53bCTNSCLMwn
New module: document-scanner.js
- PDF: extracts text from content streams, scans for PPI, converts to
sanitized plaintext for upload (PDFs can't be reliably edited in-place)
- DOCX/XLSX: extracts text from XML entries, scans for PPI
- Text files: direct string substitution
- Handles scanned/image-only PDFs gracefully (skips with message)
Fetch interceptor (content.js):
- Now intercepts FormData uploads (file uploads) in addition to JSON
- Processes each file through document scanner
- Preview mode for PDF/DOCX/XLSX: shows PPI findings with counts,
user clicks "Substitute & Upload" or "Upload Original"
- Text files substituted silently (no preview friction)
- String form fields also substituted
- 30-second auto-dismiss on preview (uploads original if no response)
Preview UI (content.css):
- Centered overlay with dark theme matching existing warning UI
- Shows original → substituted pairs for each PPI found
- File format note (e.g. "PDF will be converted to plain text")
https://claude.ai/code/session_01SWSwDfMVij53bCTNSCLMwn
Like spellcheck for privacy. Scans text as you type and paste
into chat inputs (debounced 800ms). Shows a dark floating warning
panel listing detected PPI BEFORE you hit Enter.
Each detected item has a green [+] button that instantly:
1. Generates a plausible fake value (random IP, fake address, etc.)
2. Adds it as a mapping to storage
3. Shows a checkmark to confirm
The warning disappears when you clear the text or when all
detected items have been addressed.
Three new Options toggles:
- Auto-detect unconfigured PPI (on by default)
- Offer to auto-add detected PPI (on by default)
Also adds a storage bridge (postMessage) so the page-world
content script can read/write chrome.storage through the
injector.
https://claude.ai/code/session_01Dvgwe7XMoSxnWXkih8p1Cw
Scans outbound messages AFTER all substitutions for personal data
the user forgot to configure:
- Private/public IP addresses (skips 127.0.0.1, 8.8.8.8, etc.)
- MAC addresses
- Street addresses ("123 Main St")
- GPS coordinates
- Dates (possible DOBs)
- EIN/tax IDs
- Home directory paths not caught by smart patterns
- Shell prompts (user@host)
- Git remotes (reveals username/org)
- Environment variable assignments (HOME=, USER=, etc.)
Shows a floating dark warning banner (top-right, auto-dismisses
after 15s) listing each detected item with its type, value, and
hint. Skips values already in the user's identity config.
Also shows PPI warnings in the popup Test tab and adds toggle
in Options to disable auto-detect.
https://claude.ai/code/session_01Dvgwe7XMoSxnWXkih8p1Cw
Two visual modes using the Highlight API (zero DOM modification):
- Yellow highlight: fake/substituted values the AI received.
Shows automatically in all AI responses, even without reveal mode.
"The AI sees these yellow values, not your real data."
- Terminal style (black bg, green text): your real data shown
in reveal mode. "This is YOUR data — the AI never saw this."
Uses CSS.highlights with ::highlight() pseudo-elements — no
spans, no DOM changes, no copy/paste artifacts. Falls back to
CSS class-based styling if Highlight API is unavailable.
Highlights are debounced (500ms) to handle streaming responses
without excessive reflows.
https://claude.ai/code/session_01Dvgwe7XMoSxnWXkih8p1Cw
Revealed values are now wrapped in <span class="ss-revealed">
with a blue underline and subtle blue background, making it
obvious which parts are your real data vs what the AI said.
Hover shows tooltip "Substituted value: [fake]" so you can see
what the AI actually received. Un-reveal properly removes the
spans and normalizes text nodes.
https://claude.ai/code/session_01Dvgwe7XMoSxnWXkih8p1Cw
Reveal mode (eye icon) now works as intended:
- Toggle ON: all existing responses on the page get fake→real
substitution applied immediately (paths, names, emails, etc.)
- Streaming responses are revealed in real-time as they arrive
- Toggle OFF: original text is restored from saved state
- Covers code blocks, artifacts, pre tags, and all response
containers across all supported services
- Blue floating badge shows "Reveal Mode — showing real data"
when active so user knows what they're seeing
The workflow is now: type /home/jsmith/... → Claude sees
/home/ademo/... → Claude responds with /home/ademo/... →
reveal mode shows /home/jsmith/... → user copies real path.
Also adds privacy note in popup footer: data stays in local
browser storage, no servers, no tracking, no analytics.
https://claude.ai/code/session_01Dvgwe7XMoSxnWXkih8p1Cw
Chrome Manifest V3 extension that intercepts personal data and
substitutes it with user-defined replacements before sending to
Claude.ai. Hooks fetch() in the page's main world to catch API
requests, with bidirectional substitution (real→fake on send,
fake→real on display via reveal mode).
Includes popup UI with mapping management, live test/diff view,
activity log with badge count, options page with import/export,
and Shadow DOM traversal for Claude.ai compatibility.
https://claude.ai/code/session_01Dvgwe7XMoSxnWXkih8p1Cw
Claude