Skip to main content
Glama
deenesjakoruzh
Flyingmiata-droid

paytriage

by Flyingmiata-droid
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

paytriage

Lee un error de pasarela de pago y te dice qué significa probablemente, qué comprobar a continuación y si reintentar es seguro.

Apúntalo a una línea de registro, una cadena de error o un archivo de registro completo. Responde en lenguaje sencillo en lugar de un código que tengas que buscar.

Sin dependencias. Python 3.8 y superior. Funciona como CLI, como biblioteca y como servidor MCP que un agente puede llamar.

Por qué

El personal de soporte y desarrollo pierde mucho tiempo convirtiendo response_code="51" en "la tarjeta del cliente no tenía dinero, esto no es tu error". Esa traducción vive en las cabezas de la gente y en PDFs marcados. Esto la pone en un archivo que puedes leer, buscar y corregir.

El caso de tiempo de espera agotado es el que más importa. Un timeout no es un rechazo. La transacción puede ya estar autorizada en el otro lado, así que un reintento a ciegas es cómo cobras dos veces a alguien. paytriage lo dice en voz alta cada vez.

Related MCP server: ErrorLens MCP

Instalar

git clone https://github.com/Flyingmiata-droid/paytriage.git
cd paytriage
pip install -e .

O sáltate la instalación y ejecútalo en el lugar con python -m paytriage.cli.

Úsalo desde la línea de comandos

$ paytriage 'AUTH DECLINED response_code="51" insufficient funds'

  [low] Insufficient funds  (decline)
    cause: The issuer declined the authorization because the account did not have
           enough available balance or credit.
    check: Nothing to fix on the integration side. Ask the cardholder to use
           another card or retry later.
    matched response code: 51
    safe to retry: no

Un archivo completo, resumido:

$ paytriage -f examples/gateway.log --summary

lines: 11  matched: 9  unmatched: 2

by severity:
  low: 2
  medium: 3
  high: 2
  critical: 2

by signature:
  decline.insufficient_funds: 1
  transport.tls_handshake: 1
  format.signature_mismatch: 1
  ...

Añade --json a cualquier llamada para obtener salida legible por máquina.

Úsalo como biblioteca

from paytriage import triage_line

result = triage_line("SSLError: handshake failure tlsv1 alert protocol version")

result.severity          # "critical"
result.retryable         # True
result.findings[0].label # "TLS handshake failure"
result.findings[0].fix   # what to check

Úsalo como servidor MCP

El servidor habla JSON-RPC sobre stdio y expone una herramienta, triage_payment_error, para que un agente pueda pasarle un error y recibir una salida estructurada.

python -m paytriage.mcp_server

Configuración de Claude Desktop o Claude Code:

{
  "mcpServers": {
    "paytriage": {
      "command": "python",
      "args": ["-m", "paytriage.mcp_server"]
    }
  }
}

Qué cubre

Veinticuatro firmas en seis familias:

  • Rechazos: fondos insuficientes, no honrar, tarjeta caducada, retener tarjeta, número inválido

  • Verificación: discrepancia AVS, discrepancia CVV

  • Transporte: handshake TLS, certificado caducado o no confiable, timeout, conexión rechazada o reiniciada, DNS

  • API: 401, 403, 429, 5xx

  • Formato de mensaje: parseo XML, parseo JSON, discrepancia de firma o HMAC, campo obligatorio faltante

  • Reglas de negocio: transacción duplicada, moneda no soportada, importe incorrecto, token inválido

La coincidencia se hace por código de respuesta cuando está presente, y por patrón en caso contrario. Una línea puede coincidir con más de una firma, y los resultados vuelven de peor a mejor.

Lo que no es

  • No es puntuación de fraude, ni un motor de riesgo.

  • No está conectado a ninguna pasarela. Lee texto que ya tienes y no hace llamadas de red.

  • No es exhaustivo. Es una ayuda de triaje de primera pasada. La documentación del propio emisor sigue siendo la autoridad en cualquier código específico.

  • No es un tomador de decisiones. Nada aquí debería reintentar un pago por sí solo.

Añadir una firma

Todo vive en una tabla en paytriage/signatures.py. Una firma es un patrón o un conjunto de códigos, más la causa, la solución y una severidad. Añade una entrada, añade una prueba, abre un pull request. Las correcciones a la redacción existente son tan bienvenidas como las nuevas entradas.

Pruebas

python -m unittest discover -s tests -t .

23 pruebas, sin red, sin fixtures que descargar.

Licencia

MIT. Ver LICENSE.

Available Tools

1 tool
triage_payment_errorA

Triage a payment gateway error string or log excerpt. Returns the likely cause, what to check next, a severity, and whether a retry is safe.

NameRequiredDescriptionDefault
textYesThe error message, response body, or log excerpt.
multilineNoTreat the input as many log lines instead of one.

TDQS

A4/5.0
Behavior3/5

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

No annotations are provided, so the description carries the full transparency burden. It does disclose the core behavioral trait well by enumerating return values (cause, next steps, severity, retry safety). However, it never explicitly states the operation is non-destructive/read-only (only implied by 'triage') and says nothing about edge cases such as empty input, unrecognized formats, or whether triage is heuristic versus calling an external service.

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

Conciseness5/5

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

Two tight sentences with zero filler. The purpose is front-loaded first, then the return contract is enumerated. Every clause earns its place, and the length is proportionate to the tool's simplicity.

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?

Since there is no output schema, the description correctly takes on the job of explaining return content, which it does thoroughly. Both parameters are covered by the schema, and complexity is low. The only minor gap: it could note the format expectation for text (e.g., an excerpt length limit or what 'multiline' affects) to further reduce ambiguity.

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

Parameters3/5

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

Schema description coverage is 100% — both 'text' and 'multiline' are fully documented with type, requirement, and a description. The tool description's 'error string or log excerpt' roughly mirrors what the schema already says, adding only marginal enrichment (framing 'text' as an excerpt) rather than genuinely new meaning. Baseline 3 is appropriate.

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 states a specific verb ('Triage') and resource ('payment gateway error string or log excerpt'), and enumerates exactly what it returns: likely cause, what to check next, severity, and retry safety. Though there are no siblings to distinguish from, the purpose is unambiguous and concrete.

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

Usage Guidelines4/5

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

The domain is clearly scoped to payment gateway error strings or log excerpts, which effectively conveys when to use it. With no sibling tools, there are no alternatives to exclude, but the 'when' is implicit rather than explicitly framed — no explicit statement like 'use when a payment fails with an unknown downstream reason'.

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

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 1 tool updatev0.1.0
    • First observedtriage_payment_error
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.1/5.0

Scored across 1 tool

Disambiguation5/5

With only one tool, there is no possibility of confusion or overlap. The tool's purpose is clearly singular and distinct.

Naming Consistency5/5

The single tool 'triage_payment_error' follows a clear verb_noun pattern, consistent with best practices. No inconsistency exists in a one-tool set.

Tool Count3/5

The tool count is at the thin end of the spectrum (1 tool), which feels minimal for a server, but the narrow scope of payment error triage makes it borderline acceptable.

Completeness5/5

The single tool fully addresses the stated purpose of triaging payment errors, providing cause, next steps, severity, and retry safety—no obvious gaps for this focused domain.

Maintenance

ActivityStale
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • error-broker

    amsultan2010
    A
    license
    Not graded
    quality
    D
    maintenance
    Classifies agent errors and provides clear recovery actions (retry, escalate) with configurable categories and strategies, plus MCP tools for direct agent use.
    8 npm
    1
    MIT
  • attest-mcp by attestagents

    attest-mcpofficial

    attestagents
    A
    license
    A
    quality
    D
    maintenance
    Enables AI agents to scan payment endpoints for safety, returning a letter grade (A–F) and verdict before authorizing payments.
    2
    109 npm
    MIT