Skip to main content
Glama
deenesjakoruzh
wedo911

secretscan

by wedo911
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

secretscan-mcp-server

Glama score

Un servidor MCP que analiza un diff, el contenido de un archivo o un fragmento pegado en busca de secretos confirmados por accidente, de modo que cualquier agente compatible con MCP pueda autocomprobar sus propios cambios antes de hacer commit, abrir un PR o pegar un extracto de registro en cualquier lugar. Totalmente local: sin clave de API, sin llamada de red, sin más dependencia que el SDK de MCP y Zod.

Por qué

Una credencial filtrada es una de las causas más comunes en el mundo real de compromiso de cuentas e infraestructura, y uno de los errores más fáciles de cometer en un diff grande: un console.log de depuración con una clave de API real, un archivo de configuración confirmado por accidente, un .env pegado en un chat mientras se pide ayuda. Un agente que está a punto de hacer commit, push o compartir texto está en una buena posición para detectarlo antes de que ocurra, si tiene una forma económica de comprobarlo.

Related MCP server: netallion-mcp-lite

Herramienta

scan_for_secrets

Analiza el texto comparándolo con un conjunto fijo de firmas de formato de proveedor, junto con una heurística genérica condicionada por la entropía:

Detector

Detecta

aws_access_key_id

IDs de clave de acceso de AWS (AKIA...)

aws_secret_access_key

claves secretas de AWS, cuando están etiquetadas contextualmente

github_personal_access_token / github_fine_grained_token

PATs de GitHub (ghp_..., github_pat_...)

slack_token

tokens de Slack (xoxb-..., etc.)

stripe_live_key / stripe_test_key

claves secretas/publicables de Stripe

google_api_key

claves de API de Google (AIza...)

npm_token

tokens de publicación de npm (npm_...)

private_key_block

bloques de clave privada PEM

jwt

tokens JSON Web

generic_assigned_secret

cualquier valor asignado a un nombre de variable que suene a secreto (api_key, password, token, ...) cuyo valor tenga la entropía de Shannon suficientemente alta como para parecer aleatorio en lugar de un marcador de posición como "changeme"

Los hallazgos siempre se enmascaran: AKIAIOSFODNN7EXAMPLE se notifica como AKIA************MPLE (20 chars), nunca en su totalidad. Por lo tanto, la salida de la propia herramienta se puede registrar o mostrar sin propagar más lo que haya encontrado.

Este es un escaneo de patrones fijos + heurística, no exhaustivo. Un resultado limpio significa «no coincidió ningún patrón conocido», no «definitivamente seguro»; el resultado que muestra la aplicación lo dice explícitamente.

Instalación y configuración

git clone https://github.com/wedo911/secretscan-mcp-server.git
cd secretscan-mcp-server
npm install
npm run build

Añádelo a la configuración de tu cliente MCP (por ejemplo, claude_desktop_config.json, o el .mcp.json de un proyecto para Claude Code):

{
  "mcpServers": {
    "secretscan": {
      "command": "node",
      "args": ["/absolute/path/to/secretscan-mcp-server/dist/index.js"]
    }
  }
}

Ejecutar las pruebas

npm run build
node --test tests/entropy.test.mjs tests/detectors.test.mjs

24 pruebas, incluida una para cada detector mencionado, el comportamiento de filtrado de marcadores de posición y de baja entropía del detector genérico, y una comprobación explícita de que la salida enmascarada nunca contiene el valor secreto completo.

Pruébalo sin un cliente

npx @modelcontextprotocol/inspector --cli node dist/index.js \
  --method tools/call --tool-name scan_for_secrets \
  --tool-arg text='aws_access_key_id = AKIAIOSFODNN7EXAMPLE'

(Se trata de la clave de ejemplo documentada públicamente por el propio AWS, no de una credencial real).

Licencia

MIT — consulte LICENSE.

Available Tools

1 tool
scan_for_secretsScan Text for Leaked SecretsA
Read-onlyIdempotent

Scan text (a git diff, a file's contents, a pasted snippet) for accidentally-committed secrets: cloud provider keys (AWS, Google), platform tokens (GitHub, Slack, npm), payment keys (Stripe), private key blocks, JWTs, and a generic entropy-gated check for anything assigned to a secret-sounding variable name. Use this before committing, opening a PR, or pasting logs/config anywhere -- a committed secret is one of the most common real-world causes of account and infrastructure compromise, and is easy to miss in a large diff.

Findings are always redacted (first/last few characters only) -- this tool never returns a full secret value, so its own output is safe to log or display.

Args:

  • text (string, 1-200000 chars): the text to scan.

Returns: For JSON format: { "found": boolean, "riskLevel": "none" | "medium" | "high" | "critical", "findings": [ { "detector": string, "category": string, "severity": "low"|"medium"|"high"|"critical", "line": number, "column": number, "redacted": string } ] }

Examples:

  • Use when: about to run "git commit" or "git push" -- scan the staged diff first

  • Use when: about to paste a config file, log excerpt, or error message into a chat or issue tracker

  • Don't use when: you need certainty nothing sensitive is present -- this is a fixed-pattern + heuristic scan, not exhaustive; a clean result means "no known pattern matched," not "definitely safe"

Error Handling:

  • Returns an error if text is empty or exceeds 200000 characters.

NameRequiredDescriptionDefault
textYesThe text to scan -- a git diff, a file's contents, or any pasted snippet.

TDQS

A4.7/5.0
Behavior5/5

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

Annotations already declare read-only, idempotent, and non-destructive behavior. The description adds critical behavioral context beyond this: findings are always redacted to the first/last few characters, output is safe to log, the scan is not exhaustive, and empty/oversized inputs produce errors. This meaningfully exceeds what annotations convey.

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?

The definition is well-structured with clear sections (description, args, returns, examples, error handling), front-loads the core purpose and usage guidance, and every section carries useful information. Length is justified by the tool's security-critical nature and the need to explain redaction and limitations.

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?

With no output schema, the description fully specifies the return payload structure including found, riskLevel, and findings fields. It also covers error handling, length constraints, and usage examples. For a single-parameter tool with strong annotations, this is complete and leaves no critical gaps.

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%, and the input schema already describes text as 'a git diff, a file's contents, or any pasted snippet' with length constraints. The description repeats these details, adding little new semantic meaning beyond the schema, so the baseline score of 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 opens with a specific verb and resource ('Scan text ... for accidentally-committed secrets') and enumerates the exact detector categories (cloud keys, platform tokens, payment keys, private key blocks, JWTs, entropy-gated checks). This is precise and unambiguous even without sibling tools to differentiate from.

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?

Provides explicit when-to-use scenarios ('before committing, opening a PR, or pasting logs/config anywhere') and an explicit don't-use case ('you need certainty nothing sensitive is present'). It also honestly frames the tool's limits as a fixed-pattern + heuristic scan, giving clear decision guidance.

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 updatev1.0.0
    • First observedscan_for_secrets
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

A4.6/5.0

Scored across 1 tool

Disambiguation5/5

Only one tool exists, so there is zero ambiguity or risk of misselection. The tool's purpose is clearly singular.

Naming Consistency5/5

The single tool uses a consistent verb_noun pattern ('scan_for_secrets'), which is clear and predictable.

Tool Count3/5

With only one tool, the server feels thin, but for a narrow purpose like scanning text for secrets, a single comprehensive tool is reasonable. It borders on minimal but works.

Completeness5/5

The tool covers the full scanner domain: it detects a wide variety of secrets, provides redacted output, and handles error cases. There are no obvious missing operations for its stated purpose.

Maintenance

ActivityStale
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • secret-scanner

    Baneado98
    A
    license
    A
    quality
    B
    maintenance
    Enables scanning diffs or code blobs for leaked secrets, returning a verdict with severity and masked findings, all processed locally with no data sent externally.
    1
    MIT

  • netallion-mcp-lite

    netallion
    A
    license
    Not graded
    quality
    C
    maintenance
    Scans text and files for common secrets (AWS, GitHub, etc.) and redacts them to prevent credential leakage in AI-assisted development. Runs entirely locally with no telemetry.
    24 PyPI
    MIT