Skip to main content
Glama

Preview an import

import_preview
Read-only

Dry-run parse expense data from Splitwise, Splid, Tricount or Settle Up CSV files or receipt text, showing parsed rows, member mappings, and conflicts before commit.

Instructions

Dry-run parse of expense data exported from another app (Splitwise, Splid, Tricount or Settle Up CSV) or pasted receipt text, without writing anything. Call this first, then import_commit with the same source and payload. Returns parsed (expense rows), payloadHash, memberMapping (file name to member id or null), conflicts (names still unmapped), members (active members of the flat), rejected (rows that could not be parsed) and fuzzyNames (names matched only approximately; check them). Names are matched to members automatically, and any entry you pass in memberMapping overrides that match. Rate limit: 10 previews per hour per member.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "source": "<source>",
      "payload": "<payload>"
    },
    "name": "import_preview"
  }
}

Input Schema

NameRequiredDescriptionDefault
sourceYesFormat of payload: splitwise_csv (Splitwise export), splid_csv (Splid export), tricount_csv (older Tricount exports; Tricount dropped export in 2024), settleup_csv (Settle Up export) or ocr_text (text read from a receipt or screenshot).
wgCodeNoShare code of the flat (code from wg_list). Leave it out when the account belongs to one flat.
payloadYesThe complete file or text content as one string, unmodified (max 5 MiB for CSV, 1 MiB for ocr_text, 2000 rows). Pass the identical string to import_commit.
memberMappingNoOptional overrides that map a name exactly as written in the file to a member id from members_list, or null for no member. Omit it to rely on automatic name matching, then repeat the call with corrections for entries in conflicts or fuzzyNames.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv0.3.0
    • changedInput schema / properties / source / description
      Previous value: -"Format of payload: splitwise_csv (Splitwise export), splid_csv (Splid export) or ocr_text (text read from a receipt or screenshot)."New value: +"Format of payload: splitwise_csv (Splitwise export), splid_csv (Splid export), tricount_csv (older Tricount exports; Tricount dropped export in 2024), settleup_csv (Settle Up export) or ocr_text (text read from a receipt or screenshot)."
    • changedInput schema / properties / source / enum
      Previous value: -[
      -  "splitwise_csv",
      -  "splid_csv",
      -  "ocr_text"
      -]New value: +[
      +  "splitwise_csv",
      +  "splid_csv",
      +  "tricount_csv",
      +  "settleup_csv",
      +  "ocr_text"
      +]
  2. Changed3 schema fields changedv0.1.1
    • addedInput schema / properties / memberMapping / description
      Added value: +"Optional overrides that map a name exactly as written in the file to a member id from members_list, or null for no member. Omit it to rely on automatic name matching, then repeat the call with corrections for entries in conflicts or fuzzyNames."
    • addedInput schema / properties / payload / description
      Added value: +"The complete file or text content as one string, unmodified (max 5 MiB for CSV, 1 MiB for ocr_text, 2000 rows). Pass the identical string to import_commit."
    • addedInput schema / properties / source / description
      Added value: +"Format of payload: splitwise_csv (Splitwise export), splid_csv (Splid export) or ocr_text (text read from a receipt or screenshot)."
  3. First observedv0.1.0

TDQS

A4.8/5.0
Behavior5/5

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

Beyond the readOnlyHint/destructiveHint annotations it adds material behavior: a hard rate limit of 10 previews per hour per member, the guarantee that nothing is persisted, and the rule that memberMapping entries override automatic name matching. It also pre-announces what the response contains and which parts need human-style verification (fuzzyNames).

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?

Purpose and workflow are front-loaded in the first two sentences, with mechanics and limits trailing. The long enumeration of return fields is a single dense sentence, but it is justified because no output schema exists to carry that information.

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

Completeness5/5

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

For a tool with no output schema, four parameters including a nested object, and a two-step commit workflow, the description supplies everything needed: return shape, override behavior, unresolved-name handling, and the throttling constraint. Nothing an agent would need before calling it is missing.

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; the description still adds real meaning by stressing that source and payload must be passed identically to import_commit and by explaining the override semantics of memberMapping relative to automatic matching. It does not, however, add anything about wgCode or the payload size/row limits beyond what the schema already documents.

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?

The description opens with a specific verb+resource ("Dry-run parse of expense data") and enumerates the exact accepted origins (Splitwise, Splid, Tricount, Settle Up CSV or pasted receipt text). It explicitly frames itself as the non-writing preview counterpart to import_commit, so an agent can distinguish it from any write path.

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 states the required sequencing outright: "Call this first, then import_commit with the same source and payload," and reinforces the non-mutating nature with "without writing anything." The correction loop is also spelled out — repeat the call with fixes for conflicts or fuzzyNames rather than committing blind.

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