Skip to main content
Glama

Tinplate Edge AI Safety, Security & EU Compliance

tinplate_tabular_rag_cleaner

Read-onlyIdempotent

Tabular CSV/JSON RAG Sanitizer & Formula-Injection Firewall API: Neutralize CSV formula injection, redact cell-level PII, and convert tables into RAG-ready Markdown & row chunks. (Endpoint: POST /v1/tabular-rag-cleaner)

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "tinplate_tabular_rag_cleaner"
  }
}

Input Schema

NameRequiredDescriptionDefault
csvNoRaw CSV, TSV, or semicolon-delimited table text.
rowsNoOptional structured JSON array of row objects instead of raw CSV text.
delimiterNoColumn delimiter for raw text input (default 'auto').auto
redact_piiNoRedact PII (emails, phones, IBANs, credit cards, national IDs) inside cells.
sanitize_formulasNoNeutralize CSV/spreadsheet formula injection payloads (=, +, -, @, DDE commands) by prefixing an apostrophe.
normalize_decimalsNoWhen delimiter is ';' (or auto-detected ';'), normalize standalone European decimal commas (e.g. '1,5') to dots ('1.5').
check_prompt_injectionNoScan cells for indirect prompt-injection instructions hidden in table data.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / normalize_decimals
      Added value: +{
      +  "description": "When delimiter is ';' (or auto-detected ';'), normalize standalone European decimal commas (e.g. '1,5') to dots ('1.5').",
      +  "example": false,
      +  "type": "boolean"
      +}
  2. Changed19 schema fields changed
    • addedInput schema / properties / check_prompt_injection / default
      Added value: +true
    • changedInput schema / properties / check_prompt_injection / description
      Previous value: -"Boolean flag 'check_prompt_injection' (default: true)"New value: +"Scan cells for indirect prompt-injection instructions hidden in table data."
    • addedInput schema / properties / check_prompt_injection / example
      Added value: +true
    • changedInput schema / properties / csv / description
      Previous value: -"Parameter 'csv' (e.g. \"customer,email,balance,note\\nAcme Corp,billing@acme.eu,-120.5\")"New value: +"Raw CSV, TSV, or semicolon-delimited table text."
    • addedInput schema / properties / csv / example
      Added value: +"customer_id,email,amount,notes\n101,alice@tinplate.dev,249.50,Standard Enterprise renewal\n102,bob@example.com,-15.00,\"=cmd|' /C calc'!A0\""
    • addedInput schema / properties / csv / maxLength
      Added value: +262144
    • addedInput schema / properties / csv / minLength
      Added value: +1
    • addedInput schema / properties / delimiter / default
      Added value: +"auto"
    • changedInput schema / properties / delimiter / description
      Previous value: -"Parameter 'delimiter' (e.g. \"auto\")"New value: +"Column delimiter for raw text input (default 'auto')."
    • addedInput schema / properties / delimiter / enum
      Added value: +[
      +  "auto",
      +  ",",
      +  ";",
      +  "\t",
      +  "|"
      +]
    • addedInput schema / properties / delimiter / example
      Added value: +"auto"
    • addedInput schema / properties / redact_pii / default
      Added value: +true
    • changedInput schema / properties / redact_pii / description
      Previous value: -"Boolean flag 'redact_pii' (default: true)"New value: +"Redact PII (emails, phones, IBANs, credit cards, national IDs) inside cells."
    • addedInput schema / properties / redact_pii / example
      Added value: +true
    • addedInput schema / properties / rows
      Added value: +{
      +  "description": "Optional structured JSON array of row objects instead of raw CSV text.",
      +  "items": {
      +    "additionalProperties": {
      +      "nullable": true
      +    },
      +    "type": "object"
      +  },
      +  "maxItems": 1000,
      +  "minItems": 1,
      +  "type": "array"
      +}
    • addedInput schema / properties / sanitize_formulas / default
      Added value: +true
    • changedInput schema / properties / sanitize_formulas / description
      Previous value: -"Boolean flag 'sanitize_formulas' (default: true)"New value: +"Neutralize CSV/spreadsheet formula injection payloads (=, +, -, @, DDE commands) by prefixing an apostrophe."
    • addedInput schema / properties / sanitize_formulas / example
      Added value: +true
    • removedInput schema / required
      Removed value: -[
      -  "csv",
      -  "delimiter"
      -]
  3. First observed

TDQS

A3.5/5.0
Behavior3/5

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

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, and openWorldHint=false, so the safety profile is covered. The description adds useful transformation context (formula payloads are neutralized, PII is cell-level, output is Markdown plus row chunks), but says nothing about output shape, limits (e.g. 1000 rows / 256KB), or whether flagged prompt injections are reported versus silently stripped.

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?

One dense, front-loaded sentence stating the three capabilities, followed by the endpoint in parentheses; no filler or repetition. Slightly jargon-heavy ('Firewall', 'Sanitizer & Formula-Injection Firewall API') but the operational verbs come through immediately.

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

Completeness3/5

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

For a 7-parameter, zero-required tool with no output schema, the definition covers purpose but leaves the agent without guidance on input mode selection (raw csv vs structured rows), what the response contains, or how detected prompt injections surface in the result. Annotations cover safety, so the gaps are moderate rather than severe.

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 description coverage is 100%, so the baseline is 3; the description earns a bit more by framing what the key toggles accomplish (formula neutralization and PII redaction) as the tool's core contract rather than incidental options. It still adds no semantics for the csv-versus-rows input choice, which is the primary decision an agent must make with zero required parameters.

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

Purpose4/5

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

States concrete verbs and resources: neutralize CSV formula injection, redact cell-level PII, and convert tables into RAG-ready Markdown and row chunks, plus the endpoint POST /v1/tabular-rag-cleaner. It is clear what the tool does, but it never names the closely related siblings (tinplate_pii_redactor, tinplate_rag_chunker) that an agent might otherwise choose, so differentiation is inferred rather than stated.

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

Usage Guidelines3/5

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

The phrase 'RAG-ready' and 'Firewall' imply the use case (sanitizing tables before ingestion), but there is no explicit when-to-use, when-not-to-use, or alternative routing guidance. With sibling tools that separately redact PII and chunk documents, the agent is left to guess whether this composite tool supersedes them.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources