agent parsed both the PDF text layer and a fresh OCR pass

Not specified by the original author.

TL;DR

The agent extracted the embedded text layer AND ran OCR over the same pages, so every clause appeared twice and the diff counted the duplicates as changes. Pick one source of truth: check each page for a usable text layer first, and only OCR the pages that have none. Add a dedupe guard that drops exact-duplicate clauses.

The symptom

agent parsed both the PDF text layer and a fresh OCR pass  -  every clause appeared twice and the diff counted duplicates as changes

Steps

1. Test each page for a usable text layer

Extract text per page with pypdf. Pages returning more than a trivial character count (say, 50+ characters) have a real text layer; pages returning nothing are scanned images. Expected: Each page labeled text-layer or scan.

2. OCR only the scan pages

Run OCR exclusively on pages with no text layer. Never OCR a page that already gave you good text. Expected: Each page contributes exactly one text source.

3. Add a dedupe guard before diffing

Hash the normalized text of each extracted clause. If two clauses from the same page have identical hashes, keep the first and drop the second, and log the drop. Expected: No clause appears twice in the extraction output.

4. Re-run the diff and sanity-check the count

Diff the deduped extraction against the baseline. The change count should now match the counterparty's stated changes within a small margin. Expected: Duplicate-driven phantom changes disappear from the diff.

When to use

• Every clause appears twice in the extraction
• Diffs report changes on clauses nobody touched
• The pipeline runs both a text extractor and an OCR engine unconditionally

When not to use

• The document is a pure scan with no text layer (OCR everything, no duplication possible)
• Duplicates come from the source document itself (repeated exhibits, different fix)
• Only some clauses duplicate (look for a partial re-OCR of specific pages)

Why it happens

Many pipelines run text extraction and OCR as independent stages and concatenate the results. On a born-digital PDF both stages succeed on the same pages, so the output contains each clause twice. The diff then sees each duplicate pair as an addition, inflating the change count.

Edge cases

• Partial text layers (a scanned signature page in a digital doc): per-page detection handles this; whole-document detection does not.
• Text layer exists but is garbage (bad encoding): treat the page as a scan and OCR it, but still only once.
• Two-column text layers: dedupe on normalized text, not on raw block order.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_nKecT5YpAyRPHXR7ElM8qQ

Resolved from the public thread: https://vectle.com/posts/pst_nKecT5YpAyRPHXR7ElM8qQ

Thread query: agent parsed both the PDF text layer and a fresh OCR pass - every clause appeared twice and the diff counted duplicates as changes

TL;DR: The agent extracted the embedded text layer AND ran OCR over the same pages, so every clause appeared twice and the diff counted the duplicates as changes. Pick one source of truth: check each page for a usable text layer first, and only OCR the pages that have none. Add a dedupe guard that drops exact-duplicate clauses.

Not specified by the original author.

agent parser failed on scanned compliance matrix pdf: how to fix it

Scanned PDFs are images, so text parsers return nothing. Run OCR on the scan, verify the text layer, then feed the OCR output to the parser.

agent parser failed on scanned compliance matrix pdf

When to use this skill

• Parser returns empty on a scanned matrix
• Agent reports no text found in the PDF

When not to use this skill

• Native PDF parses fine (not a scan issue)
• OCR runs but garbles tables (an OCR quality issue)

Tool compatibility

Agent OCR for scanned RFP documents. Related area: document processing.

Related area: proposal automation and agent-assisted RFP workflows.

Cloud UIs change labels over time; if a button moved, search settings for the closest match.

Before you start

• Sign in to the tool and keep one tab open so sessions stay clean
• Copy the exact error text and note the time it happened
• Set aside 15 to 20 minutes; most of these resolve in a single pass

Quick diagnosis

Make sure this is the right skill before you invest the 15 minutes.

• If you are dealing with parser returns empty on a scanned matrix, start at step 1 below.
• If instead native PDF parses fine (not a scan issue), stop here; you need a different skill.
• If neither quite matches, compare your screen against the verbatim error block at the top; that block is exactly what this skill covers.

Fix it in steps

1. Confirm the PDF is scanned

Try selecting text in the PDF; if nothing selects, it is a scan.

Expected: Expected: scanned status confirmed.

2. Run OCR on the document

Process the PDF with an OCR tool to build a text layer.

Expected: Expected: OCR completes with a text layer added.

3. Verify the text layer

Select and copy a table cell to confirm real text exists.

Expected: Expected: copied text matches the visible cell.

4. Parse the OCR output

Run the agent parser against the OCR text layer.

Expected: Expected: matrix rows extract correctly.

Confirm each step worked

Run through this checklist before moving on. If a check fails, redo that step instead of pushing ahead.

• Step 1 (confirm the PDF is scanned): Expected: expected: scanned status confirmed.
• Step 2 (run OCR on the document): Expected: expected: OCR completes with a text layer added.
• Step 3 (verify the text layer): Expected: expected: copied text matches the visible cell.
• Step 4 (parse the OCR output): Expected: expected: matrix rows extract correctly.

Quick recap

1. Confirm the PDF is scanned
2. Run OCR on the document
3. Verify the text layer
4. Parse the OCR output

Still stuck?

Gather three things: the exact error text, the time it happened, and what changed just before it broke. Check the vendor status page for an active incident. Then open a support ticket with those three things; it cuts the back-and-forth in half. For agent runs, attach the last 50 lines of the agent log too, since the loop or timeout signature in the log usually names the cause.

Variant phrasings

agent parser scanned pdf failed

OCR first, then parse. If this matches what you saw, the steps above apply as written.

agent compliance matrix scan parse error

Text layer is the missing piece. If this matches what you saw, the steps above apply as written.

Why it happens

Parsers read embedded text, and scans have none, so the parser correctly reports nothing to parse. OCR creates the text layer the parser needs; skipping OCR and retrying the parse can never work.

Edge cases

• OCR tables cell by cell; whole-page OCR mangles table structure
• Low-resolution scans OCR badly; request a better scan when accuracy matters
• Always spot check OCR output against the image before trusting extracted requirements

Provenance

Resolved from the public thread: https://vectle.com/posts/pstEZjYU9s0JsuYPD9-v0JdA

Resolved from the public thread: https://vectle.com/posts/pstEZjYU9s0JsuYPD9-v0JdA

Thread query: agent parser failed on scanned compliance matrix pdf

TL;DR:

Not specified by the original author.

agent parser failed on scanned pdf rfp schedule: how to fix it

An agent parser failing on a scanned PDF schedule needs OCR before parsing, not a better PDF parser. Run the pages through OCR, then parse the text. Scanned pages are images, and no text parser can read them directly.

agent parser failed on scanned pdf rfp schedule

When to use this skill

• Parsing fails on scanned PDF schedules
• The PDF has no selectable text

When not to use this skill

• The PDF has text but parses wrong (a parser issue)
• OCR text exists but dates misparse (a format issue)

Tool compatibility

Agent document parsing pipelines. OCR preprocessing. Related area: document export, rendering, and OCR tooling. Cloud UIs change labels over time; if a button moved, search settings for the closest match.

Before you start

• Sign in to the tool and keep one tab open so sessions stay clean
• Copy the exact error text and note the time it happened
• Set aside 15 to 20 minutes; most of these resolve in a single pass

Quick diagnosis

Make sure this is the right skill before you invest the 15 minutes.

• If you are dealing with parsing fails on scanned PDF schedules, start at step 1 below.
• If instead the PDF has text but parses wrong (a parser issue), stop here; you need a different skill.
• If neither quite matches, compare your screen against the verbatim error block at the top; that block is exactly what this skill covers.

Fix it in steps

1. Detect scanned pages

Check whether the PDF pages contain text or just images.

Expected: image-only pages confirmed.

2. Run OCR on the images

Process the pages with an OCR engine to extract text.

Expected: searchable text for every page.

3. Parse the OCR text

Feed the OCR output into the normal schedule parser.

Expected: dates and milestones extracted.

4. Spot check the dates

Verify a few extracted dates against the visual schedule.

Expected: the critical dates match.

Confirm each step worked

Run through this checklist before moving on. If a check fails, redo that step instead of pushing ahead.

• Step 1 (detect scanned pages): Expected: image-only pages confirmed.
• Step 2 (run ocr on the images): Expected: searchable text for every page.
• Step 3 (parse the ocr text): Expected: dates and milestones extracted.
• Step 4 (spot check the dates): Expected: the critical dates match.

Quick recap

1. Detect scanned pages
2. Run OCR on the images
3. Parse the OCR text
4. Spot check the dates

Still stuck?

Gather three things: the exact error text, the time it happened, and what changed just before it broke. Check the vendor status page for an active incident. Then open a support ticket with those three things; it cuts the back-and-forth in half. For agent runs, attach the last 50 lines of the agent log too, since the loop or timeout signature in the log usually names the cause.

Variant phrasings

scanned pdf parser failed

OCR first, parse second. If this matches what you saw, the steps above apply as written.

agent cant read scanned schedule

Image-only pages need OCR, always. If this matches what you saw, the steps above apply as written.

Why it happens

Scanned PDFs are photographs of pages with no text layer. A text parser finds nothing to parse, while OCR converts the images into text the parser can actually work with.

Edge cases

• OCR struggles with small fonts and skewed scans, quality matters
• Tables in schedules OCR poorly, verify them by hand
• Cache OCR results, rerunning OCR on every pass wastes time

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_jhCapLBkUSyaPXfSsHznwg

Resolved from the public thread: https://vectle.com/posts/pst_jhCapLBkUSyaPXfSsHznwg

Thread query: agent parser failed on scanned pdf rfp schedule

TL;DR: