Die Oberfläche ist lokalisiert; Skill-Beschreibungen und -Dokumente werden per KI automatisch in Ihre Sprache übersetzt. Code, Befehle, Pfade und Fachbegriffe bleiben im Original, maßgeblich ist das Dokument des Autors.
Skills entdecken
Daten und Forschung · SKILL.md

brightdata-debug-bundle

brightdata-debug-bundle

Assemble a Bright Data support bundle that preserves diagnostic value while excluding credentials, target data, and secret-bearing URLs. Use when escalating a persistent proxy, Browser API, snapshot, or delivery incident. Trigger with: "build a Bright Data debug bundle", "prepare evidence for Bright Data support", "redact a Bright Data incident".

Independent GitHub snapshot · Automated classification · Original documentation governs

Skill-Dokumentation

GitHub

name: brightdata-debug-bundle description: 'Assemble a Bright Data support bundle that preserves diagnostic value while excluding credentials, target data, and secret-bearing URLs. Use when escalating a persistent proxy, Browser API, snapshot, or delivery incident. Trigger with: "build a Bright Data debug bundle", "prepare evidence for Bright Data support", "redact a Bright Data incident".' allowed-tools: Read, Grep, Write, Edit, Bash(python:*) version: 2.0.0 argument-hint: "[run-receipt]" model: inherit effort: high license: MIT author: Jeremy Longshore jeremy@intentsolutions.io tags:

  • saas
  • web-data
  • bright-data
  • debug-bundle
  • operations compatibility: 'Requires an approved Bright Data account or offline fixtures, current Bright Data documentation, and an authorized public-data collection purpose'

Bright Data Redacted Support Bundle

Overview

Build a deterministic manifest of configuration shape, timestamps, versions, provider codes, and operation identifiers. Default-deny all raw headers and payloads, then include only explicitly allowlisted diagnostic fields.

Prerequisites

  • A run receipt and named incident owner
  • A private output directory with a retention deadline
  • A redaction policy covering URLs, credentials, cookies, headers, and target data

Instructions

Step 1: Inventory inputs

Read the receipt and Grep candidate logs for secret names, proxy URLs, Bearer values, cookies, query strings, and target content before copying anything.

Step 2: Create the manifest

Write a small metadata file with product, zone alias, client/runtime version, UTC window, operation ID, state, and current provider error fields.

{"product":"web-scraper-api","zone_alias":"production-redacted","operation_id":"s_redacted","provider_code":"client_10000","raw_payload_included":false}

Step 3: Redact and validate

Use Edit only to remove disallowed material from the private bundle. Use Bash(python:*) to scan for configured secret values and forbidden keys; fail closed if any match remains.

Step 4: Seal the evidence

Record sorted file hashes, owner, purpose, recipient, and deletion time. Transmit only through the approved private support channel; never attach it to a public issue.

Tool Discipline

Use Read and Grep for discovery and pre-copy review. Use Write and Edit only inside the exact private bundle path. Use Bash(python:*) for local redaction and checksum checks; it must not upload the bundle or call Bright Data.

Output

  • Private manifest plus allowlisted redacted diagnostics
  • Zero-secret scan and sorted file-hash receipt
  • Named recipient, retention deadline, and deletion owner

Examples

Include Proxy-Status, x-brd-err-code, a rounded UTC window, and an opaque snapshot ID. Exclude proxy username/password, API key, target URL query, response body, cookies, and collected records.

Error Handling

FailureMeaningResponse
Secret scan finds a valueRedaction is incompleteQuarantine the bundle and rebuild from the allowlist
Raw target content is required to reproduceSupport request exceeds the safe defaultObtain data-owner approval and use a separate controlled transfer
No operation identifier existsEvidence cannot be correlatedReproduce once against an approved test target

Resources

Gespeicherte Skills