Skip to main content
Glama
lokesh-sparrow

PNPC-MCP-Tally-Prime

create_vouchers_batch

Posts many TallyPrime vouchers in one call, verifies each chunk after posting, and reports which vouchers landed for safe partial retries.

Instructions

Post many vouchers (Journal, Payment, Receipt, etc. — same shape as create_voucher, one entry per voucher) in one call, for bulk work where posting one at a time is impractical (e.g. hundreds of balance-confirmation adjustment journals). Internally splits into chunks (default 10, max 50 — Tally's own gateway has a 10-second timeout per request, and a bigger single import risks silently exceeding it) and, after each chunk posts, runs ONE bulk query to confirm exactly which vouchers in that chunk actually exist in Tally now — this always runs, even if the post itself errored or timed out, because a client-side timeout does not mean Tally didn't still create some or all of them server-side; retrying without checking first risks posting duplicates. Verification matches each voucher back by (voucherType, narration) — NOT voucherNumber, confirmed live to be unreliable even when set explicitly (Tally silently reassigned its own auto-series number instead for a Journal voucher); reference was tried next and also came back empty on read. narration is what's confirmed live to persist exactly as sent, so it's REQUIRED and must be non-empty on every voucher, and must be unique per voucherType across the whole batch — two vouchers sharing the same (voucherType, narration) can't be told apart afterward and are rejected up front as invalid rather than silently mismatched (e.g. embed each row's own reference/id in the narration text). The result names exactly which vouchers were confirmed and which were not, per chunk, so a partial failure can be retried by resending only the ones that didn't land — never blindly resend the whole batch. Does not go through preview_write/confirm_write given the scale involved; sending a small batch (2-3 vouchers) first to confirm shape/ledger names before a large run is recommended.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
vouchersYesOne entry per voucher, same fields as create_voucher — narration is required here and must be unique per voucherType across the batch (see the tool description).
chunkSizeNoVouchers per internal request. Defaults to 10. Keep at or below 50 — Tally's gateway has a 10-second timeout per request and larger imports take proportionally longer server-side.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv1.13.1
    • addedInput schema / properties / vouchers / items / properties / creditDescription
      Added value: +{
      +  "description": "Same as debitDescription, for the credit leg.",
      +  "items": {
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedInput schema / properties / vouchers / items / properties / debitDescription
      Added value: +{
      +  "description": "Additional description lines for the debit leg (Tally's 'Additional Descriptions for Ledgers'), one array entry per line. Optional.",
      +  "items": {
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedInput schema / properties / vouchers / items / properties / entries / items / properties / description
      Added value: +{
      +  "items": {
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
  2. Addedv1.13.0

TDQS

A4.8/5.0
Behavior5/5

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

Goes well beyond the annotations: it discloses internal chunking (default 10, max 50, Tally's 10s gateway timeout), that a verification query always runs even after an error/timeout, why voucherNumber/reference are unusable for matching and narration is used instead, that duplicate (voucherType, narration) pairs are rejected up front, and the partial-failure/retry contract. This is consistent with idempotentHint=false — it explains the exact non-idempotency risk and how to avoid duplicate posts.

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?

Front-loaded with the purpose and the practical recommendation at the end; every sentence carries operational information and little is filler. It is a dense run-on with stacked parentheticals rather than clearly separated sections, which makes it harder to scan than it could be.

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 complex, no-output-schema batch mutation, the description covers the essentials an agent needs: the chunking/timeout model, verification and matching semantics, dedupe rejection rules, and the shape of the result (per-chunk confirmed vs. not-confirmed list) plus the retry-only-failed contract. Nothing material is left ambiguous.

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 genuine semantics: why narration is REQUIRED and must be unique per voucherType across the batch (the verification key), and the rationale for the chunkSize cap against the 10-second timeout. It also clarifies the vouchers entries share create_voucher's shape. Slight overlap with schema text keeps it from a 5.

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?

States a specific verb and resource ("Post many vouchers ... in one call") and explicitly scopes it as bulk work versus the sibling create_voucher's one-at-a-time posting. It also distinguishes itself from preview_write/confirm_write, so an agent can route correctly without opening any schema.

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?

Gives when-to-use (bulk work where posting singly is impractical, e.g. hundreds of adjustment journals), when-not (does not go through preview_write/confirm_write at this scale), and a concrete operational recommendation (send a 2-3 voucher batch first to validate shape/ledger names). Alternatives are named explicitly.

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

Deploy Server

Other Tools