Skip to main content
Glama

Search documents

paperless_search_documents
Read-only

Find documents in paperless-ngx by combining full-text search with structured filters like tags, correspondents, and date ranges.

Instructions

Find documents in paperless-ngx. Combine full-text search with structured filters. query uses the full-text index and supports the advanced syntax: invoice AND (acme OR globex), type:invoice tag:unpaid, correspondent:university, created:[2005 to 2009], added:yesterday, produ*name, custom_fields.name:"Contract Number", custom_fields.value:policy, notes.note:reminder. Date keywords: today, yesterday, "previous week", "this month", "previous month", "this year", "previous year", "previous quarter". Matching is word-order-independent and accent-insensitive. Use title_contains/content_contains instead for plain substring matching, or more_like_id to find documents similar to a known one. Tag/correspondent/type filters take ids — get them from paperless_list_objects. Returns names resolved, not raw ids.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
pageNo1-based page number.
queryNoFull-text query using the advanced search syntax described above.
orderingNoSort field, `-` prefix for descending. Common values: `created`, `-created`, `added`, `-added`, `modified`, `title`, `archive_serial_number`, `correspondent__name`, `document_type__name`, `num_notes`, `page_count`.
tags_allNoOnly documents carrying every one of these tag ids.
tags_anyNoOnly documents carrying at least one of these tag ids.
is_taggedNotrue = only tagged documents, false = only untagged.
mime_typeNoSubstring match on mime type, e.g. `pdf`, `image/`.
owner_idsNoRestrict to documents owned by these user ids.
page_sizeNoResults per page (max 100).
tags_noneNoExclude documents carrying any of these tag ids.
added_afterNoInclusive lower bound on when it was added, `YYYY-MM-DD`.
is_in_inboxNotrue = only documents still carrying an inbox tag.
added_beforeNoInclusive upper bound on when it was added, `YYYY-MM-DD`.
more_like_idNoReturn documents similar to this document id. Cannot be combined with `query`.
created_afterNoInclusive lower bound on the document date, `YYYY-MM-DD`.
extra_filtersNoEscape hatch for any other documented query parameter, e.g. `{"checksum__iexact":"…"}`.
snippet_charsNoCharacters of document text to include per result when there is no search highlight.
created_beforeNoInclusive upper bound on the document date, `YYYY-MM-DD`.
title_containsNoCase-insensitive substring match on the title.
content_containsNoCase-insensitive substring match on the extracted text.
storage_path_idsNoRestrict to these storage path ids.
correspondent_idsNoRestrict to these correspondent ids.
document_type_idsNoRestrict to these document type ids.
custom_field_queryNoJSON custom-field query, passed through verbatim. Examples: `["due","range",["2024-08-01","2024-09-01"]]`, `["customer","exact","bob"]`, `["answered","exact",true]`, `["foo","exists",false]`. Operators: exact, in, isnull, exists (all types); icontains/istartswith/iendswith (text); gt/gte/lt/lte/range (number, date); contains (document link).
archive_serial_numberNoExact archive serial number.
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only provide readOnlyHint and openWorldHint, so the description carries the burden of behavioral disclosure, and it delivers: it documents advanced query syntax, date keywords, word-order-independent and accent-insensitive matching, and the fact that returned names are resolved rather than raw ids. It does not contradict annotations, though it omits details about result response structure.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long, but the length is justified by the tool's 25-parameter search surface. It front-loads the purpose, then moves through syntax, alternative routes, and return behavior without filler. Each block earns its place, though it could be slightly more scannable.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex search tool with no output schema, the description covers the ambiguous parts: query syntax, date keywords, matching semantics, id-based filters, and resolved-name return behavior. It appropriately relies on the schema for the many structured filter parameters. The main missing piece is a fuller description of the result payload shape.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds substantial value beyond the schema: it teaches the full-text query syntax with rich examples, clarifies that tag/correspondent/type filters consume ids, and points to paperless_list_objects as the id source. This materially helps an agent use the parameters correctly.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Description opens with 'Find documents in paperless-ngx' — a specific action and resource — and immediately differentiates itself from retrieval/download siblings. It also names the closest semantic alternatives ('title_contains', 'content_contains', 'more_like_id'), so the tool's unique search scope is unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It explicitly tells agents when to prefer alternatives: use title_contains/content_contains for plain substring matching or more_like_id for similarity search. It also directs the agent to paperless_list_objects for id-based filters, preventing the common mistake of passing names instead of ids.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/patrickcylai/paperless-ngx-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server