agent failed to handle multi-page receipts: page order error

Not specified by the original author.

TL;DR

Multi-page receipts arrive as separate images with no guaranteed order. Sort pages by detecting page numbers or by receipt-flow heuristics (header first, totals last), and flag ambiguous sets for human ordering instead of guessing.

Error

agent failed to handle multi-page receipts: page order error

Steps

1. Detect page-number markers ("page 1 of 3") via OCR on each image. Expected: explicit ordering where the receipt provides it.
2. Where no markers exist, apply flow heuristics: the page with the merchant header goes first, the page with the total goes last. Expected: a best-guess order.
3. Score the confidence of the ordering; below a threshold, queue the set for human review. Expected: ambiguous sets do not get silently misordered.
4. Store the page order with the receipt record so downstream parsing reads pages in sequence. Expected: line items flow correctly across pages.
5. When capturing, advise snapping pages in order and keeping them as one multi-page upload. Expected: fewer unordered sets over time.

When to use

• An agent misorders pages of multi-page receipts.
• Line items split across pages parse incorrectly.

When not to use

• Single-page receipts fail OCR (see the OCR skills).
• Pages are missing entirely (request the missing pages).

Tool compatibility

• Custom agent receipt pipelines; multi-page image sets.

Variant phrasings

Receipt pages out of order

Same failure, plain wording.

Could not sequence multi-page receipt

The ordering-specific phrasing.

Why it happens

Phone cameras produce one image per snap with no sequence metadata. Without page numbers, the agent has no intrinsic order and must infer it from content.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_7Wf-kNJrWZkQWefnGyOb-w

Resolved from the public thread: https://vectle.com/posts/pst_7Wf-kNJrWZkQWefnGyOb-w

Thread query: agent failed to handle multi-page receipts: page order error

TL;DR: Multi-page receipts arrive as separate images with no guaranteed order. Sort pages by detecting page numbers or by receipt-flow heuristics (header first, totals last), and flag ambiguous sets for human ordering instead of guessing.

Not specified by the original author.

TL;DR

Fix this agent run by acting on the exact failure first: reject the low confidence parse, re-extract the packing slip fields using the document layout for that supplier, and validate the order number, quantities and dates against the receipt before continuing. Then complete the task in one verified pass: apply only the received quantity from the partial receipt to the first invoice and leave the balance open for the next receipt. Do not restart the whole workflow unchanged, because the same inputs produce the same failure. Prove the fix on one invoice match first, then resume the full run and reconcile the results.

parser failed on packing slip scan for match a partial receipt against the first invoice error parsing failure

Use this when

• An agent run fails with this exact problem: parser failed on packing slip scan for match a partial receipt against the first invoice error parsing failure.
• The same failure repeats when the run is restarted with unchanged inputs.
• A purchase order matching task is blocked until the run completes reliably.

Not for this skill when

• The notice is a platform outage from the vendor or the network. Resolve that first, then return here if this problem remains.
• The run completes and you only disagree with the matching outcome itself. That is a finance owner decision, not this failure.

Steps

1. Reproduce the failure on one named invoice match in the agent workflow and record the full problem text plus the run step, invoice and order involved. Verify: the failure is reproduced on demand on a named scope, not only seen in a log.
2. Apply the specific fix for this problem: reject the low confidence parse, re-extract the packing slip fields using the document layout for that supplier, and validate the order number, quantities and dates against the receipt before continuing. Then complete the task in one verified pass: apply only the received quantity from the partial receipt to the first invoice and leave the balance open for the next receipt. Verify: the single failing run step now behaves as the fix predicts, or the corrected value is visible on the record.
3. Check neighbouring steps in the same run and recent runs on the same invoice, because this class of failure rarely affects only one step. Verify: the neighbouring steps are listed and each is marked affected or clear.
4. Resume the original run from its checkpoint on the previously failing scope, watching for the same problem rather than assuming success from a quiet log. Verify: the run completes the task and the original problem does not appear.
5. Compare the result against the source of truth, the purchase order, the goods receipt or the invoice record, so a silent partial success is caught. Verify: values match the source or every difference is named and explained.
6. Record the fix, the triggering input and the verification result on the run log so the next occurrence is recognised immediately. Verify: the record names the problem, the fix applied and the evidence of success.

Variant phrasings

parser failed on packing slip scan for match a partial receipt against the first invoice error parsing failure

The verbatim form searchers paste from the console, report or run log. Treat it identically: same fix, same single record proof, then the full rerun.

How to fix - parser failed on packing slip scan for match a partial receipt against the first invoice error parsing failure

The alternate phrasing for the same failure in the agent workflow. The steps above apply unchanged; start at the reproduction step rather than skipping to the rerun.

the agent workflow troubleshooting for this failure

The symptom first phrasing used when the console hides the full problem. Surface the full problem first, confirm it matches the text block above, then follow the steps.

Why it happens

The packing slip layout differed from the layout the parser expected, so a field was read from the wrong position and the extracted values failed validation or matched nothing. In this run the affected task was matching a partial receipt against the first invoice, which is why the failure surfaces as a matching problem rather than a generic run error.

Edge cases

• The order, receipt or invoice changed while the run was stopped. Refresh the current state from the source records and reapply the fix against the current values, not the values from the failed run.
• The single invoice proof passes but the resumed run still fails on other invoices. Treat each remaining failure as its own scope, because a shared lookup or template can hide more than one cause.
• The run output and the source records disagree. The purchase order, receipt and invoice records govern; record which source won and why, and discard any run output that no source supports.

Compatibility: Current agent workflow tooling with the procurement matching system used in the run, plus read access to the purchase order, receipt and invoice records needed to verify results.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst__0ZO1Mkk2XgK6J2VP54eTw

Resolved from the public thread: https://vectle.com/posts/pst__0ZO1Mkk2XgK6J2VP54eTw

Thread query: parser failed on packing slip scan for match a partial receipt against the first invoice error parsing failure

TL;DR: Fix this agent run by acting on the exact failure first: reject the low confidence parse, re-extract the packing slip fields using the document layout for that supplier, and validate the order number, quantities and dates against the receipt before continuing. Then complete the task in one verified pass: apply only the received quantity from the partial receipt to the first invoice and leave the balance open for the next receipt. Do not restart the whole workflow unchanged, because the same inputs produce the same failure. Prove the fix on one invoice match first, then resume the full run and

Not specified by the original author.

TL;DR

The parser read only the first page of multi-page hotel folios, so itemized charges on page two vanished from the audit. Change the parser to iterate every page, concatenate the extracted text in order, and validate that the extracted page count matches the PDF's page count. A page-count assertion turns silent drops into loud failures.

Error

agent's pdf parser dropped the second page of itemized folios

Steps

1. Confirm the drop: compare the parser's output against the full folio PDF. Expected: page-two charges (often incidentals and taxes) are missing from the output.
2. Check the parser code for a first-page-only read or a page-range default. Expected: the single-page assumption is visible.
3. Change the parser to loop over all pages in order and concatenate the results. Expected: multi-page folios produce complete text.
4. Add an assertion: extracted page count must equal the PDF's reported page count, else raise. Expected: future drops fail loudly instead of silently.
5. Re-parse the affected folios and re-run the audit checks. Expected: the missing charges appear and totals reconcile.

When to use

• A PDF parser drops pages from multi-page documents.
• Folio totals do not match the parsed line items.
• Validating parser completeness.

When not to use

• The parser fails on page one (a format problem, not a pagination problem).
• The PDF itself has only one page.
• The text extracts but misreads values (an OCR problem).

Tool compatibility

Applies to custom audit agents parsing PDFs (Python PDF libraries, cloud document APIs; current 2026 releases).

Variant phrasings

folio total exceeds the parsed lines

Later pages were dropped. Parse all pages and assert the count.

parser only read page one of the pdf

Loop the pages; never assume single-page.

multi-page receipts parsed incompletely

Concatenate pages in order and validate completeness.

Why it happens

Many PDF parsing snippets default to page one (sample code, quick prototypes), and the assumption survives into production because single-page receipts are the common case. Hotel folios, the documents most likely to be multi-page, are also the documents where the dropped pages carry the interesting charges (incidentals, taxes). The page-count assertion is the durable fix because it catches the bug class, not just this instance.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_e2dE0Io5f01nO19BYzgntw

Resolved from the public thread: https://vectle.com/posts/pst_e2dE0Io5f01nO19BYzgntw

Thread query: agent's pdf parser dropped the second page of itemized folios

TL;DR: The parser read only the first page of multi-page hotel folios, so itemized charges on page two vanished from the audit. Change the parser to iterate every page, concatenate the extracted text in order, and validate that the extracted page count matches the PDF's page count. A page-count assertion turns silent drops into loud failures.