brd-browser-debug

A debugging tool for Bright Data Scraping Browser sessions, which provide remote browsers for automated web scraping. It examines live session data to diagnose errors, bandwidth use, captchas, and recurring patterns.

In plain words
What is it for?
Use it to review recent sessions, diagnose errors, analyze bandwidth, report captchas, and detect patterns through the Browser Sessions API.
Why use it?
It helps explain failed scraper runs and session problems without manually piecing together API responses and recent history.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/brightdata/skills/brd-browser-debug
Any agent
npx skills add brightdata/skills --skill brd-browser-debug
Clone the repo
git clone --depth 1 https://github.com/brightdata/skills

Made for: Claude Code, Codex.

Per session 209 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,789 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00209 $0.01789
Opus 5 $0.00105 $0.00894
Sonnet 5 $0.00042 $0.00358
Haiku 4.5 $0.00021 $0.00179

Measured yesterday against content hash 1375fd4cff52, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

brd-browser-debug scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/brd-browser-debug/SKILL.md · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Bright Data — Browser Session Debugger

Diagnose Bright Data Scraping Browser sessions using the Browser Sessions API. Fetches live session data and performs smart triage: error diagnosis, bandwidth analysis, captcha reporting, and pattern detection across recent sessions.

Setup

Set your API key:

export BRIGHTDATA_API_KEY="your-api-key"

Get a key from Bright Data Dashboard → API Tokens.

No zone configuration needed — zone is returned as a field in session data.

Usage

List & triage recent sessions

Invoked as /brd-browser-debug with no arguments.

API reference: GET /browser_sessions

Fetching sessions

Start with a single call using limit=100 (the maximum) sorted by most recent:

GET https://api.brightdata.com/browser_sessions?limit=100&sort=timestamp&order=desc
Authorization: Bearer $BRIGHTDATA_API_KEY

Pagination: The response includes total, has_more, and next_offset. If has_more is true and the analysis requires more data (e.g. bandwidth outlier detection needs a larger sample), fetch the next page using offset=<next_offset>. Continue until you have enough data or has_more is false.

Available filters — apply when the user specifies a scope:

  • status=failed|finished|running — narrow to a specific session state
  • api_name=<zone> — filter to a specific Bright Data zone
  • target_url=<domain> — filter by target domain (e.g. ksp.co.il)
  • start_date / end_date — ISO 8601 datetime range
  • sort=timestamp|duration|bandwidth with order=asc|desc

If the user asks about a specific zone, date range, or domain — apply the relevant filter rather than fetching all sessions and filtering client-side.

Triage steps
  1. Present a health summary: total from the response, counts of finished / failed / running.
  2. Most recent session — always highlight it regardless of status (same detail level as single-session mode).
  3. Failed sessions — for each failure: session ID, timestamp, duration, bandwidth, then reason about the cause using the signals in the Diagnosing Failed Sessions section below.
  4. Pattern detection — if 3+ sessions share the same error.code, call it a systemic issue:

    "3 sessions failed with custom_headers — you are overriding a header Bright Data forbids. Remove page.setExtraHTTPHeaders() from your code."

  5. Bandwidth outliers — group sessions by target_url domain. For each domain with 3+ sessions, calculate the median bandwidth. Flag any session whose bandwidth exceeds 2× the median for that domain as an outlier, and note if it was a failed session that burned unusually high bandwidth before dying.
  6. Captcha activity — report how many sessions hit captchas and whether they were solved.
  7. Close with a one-line verdict: the most important finding and the most impactful fix.

Read the full file on GitHub · 110 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 110 lines · 209 tokens per session scan A 1375fd4cff52

Subscribe to this mod's changes

brd-browser-debug is a skill published in the GitHub repository brightdata/skills (256 stars, last pushed 2d ago), licensed MIT. It adds 209 tokens to every session and 1,789 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens