CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

browser-cdp

Use this skill when you need to control a Chrome browser via CDP (Chrome DevTools Protocol) to reuse existing login sessions. Covers: launching Chrome in debug mode, opening URLs, waiting for page load, evaluating JavaScript, taking snapshots, and extracting auth tokens. Trigger phrases: browser automation, CDP, agent-browser, 浏览器操作, 操作浏览器, Chrome CDP, 复用登录态, extract token from browser.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An operationally excellent body: every command is copy-paste executable, the setup flow is a validated decision tree with an explicit consent checkpoint for destructive operations, and error recovery (exit codes, PID verification, timeouts, retries) is covered. The only improvement areas are mild repetition of the consent flow and keeping the full FAQ/limitations tables inline in an already long single file.

DimensionReasoningScore

Conciseness

The body is dense and operational — no space is spent explaining what CDP or Chrome is, and every section carries commands or decision rules. Minor trimming is possible: the consent flow for killing Chrome is restated in the preamble, the dedicated section, the exit-code notes, and the FAQ, which is mild redundancy even if deliberate for a safety-critical rule. This fits anchor 4 (efficient, minor instances that could be trimmed) rather than anchor 5's every-token-earns-its-place.

4 / 5

Actionability

Guidance is fully executable throughout: exact commands (`node {SKILL_DIR}/scripts/setup-cdp-chrome.js 9222 --detect-only`, `agent-browser --cdp 9222 eval -b "..."`), documented exit codes (0/1/2/3), copy-paste timeout wrappers for both PowerShell and bash, and per-problem fixes in the FAQ. This matches anchor 5: copy-paste ready commands covering the common cases.

5 / 5

Workflow Clarity

The startup flow is an explicit ordered decision tree (detect-only first, then branch on CDP_STATUS/CHROME_RUNNING), and the destructive Chrome-termination path has a full validate-confirm loop: dry-run to count affected processes, inform the user, re-run with --yes only after consent, with exit code 3 as the non-TTY guard. PID verification before killing processes and retry-after-timeout guidance add error-recovery feedback loops, matching anchor 5; the destructive-operation cap of 3 does not apply because validation checkpoints are present.

5 / 5

Progressive Disclosure

Sections are well-organized with clear headers, and the single bundle file (`scripts/setup-cdp-chrome.js`) is a real, correctly-pathed, one-level-deep reference invoked directly in commands. However, the ~165-line body keeps everything inline — the FAQ and known-limitations tables are content that could plausibly live in separate reference files — which is minor organization gap territory (anchor 4) rather than the cleanly split anchor 5.

4 / 5

Total

18

/

20

Passed

Description

88%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description that clearly states what the skill does, when to use it, and concrete trigger phrases in both English and Chinese. The only weaknesses are a couple of missing natural trigger variants and the slightly broad "browser automation" trigger that could collide with general web-automation skills.

DimensionReasoningScore

Specificity

The description enumerates six concrete actions — "launching Chrome in debug mode, opening URLs, waiting for page load, evaluating JavaScript, taking snapshots, and extracting auth tokens" — which comprehensively covers the skill's capabilities with no vague filler. This matches the anchor for multiple specific concrete actions with comprehensive coverage; it exceeds anchor 4 because there are no notable gaps in the action list.

5 / 5

Completeness

It explicitly answers both questions: what ("control a Chrome browser via CDP (Chrome DevTools Protocol) to reuse existing login sessions" plus the enumerated actions) and when ("Use this skill when you need to..." plus a dedicated concrete trigger-phrase list). This mirrors the anchor-5 example structure exactly; anchor 4 would require the 'when' to be less explicit, which it is not.

5 / 5

Trigger Term Quality

Trigger phrases include "browser automation, CDP, agent-browser, 浏览器操作, Chrome CDP, 复用登录态, extract token from browser" — good natural-term coverage with synonyms in both English and Chinese. A few natural phrases users might say are missing (e.g., "reuse login", "browser session", "extract cookies"), so it sits at anchor 4 rather than the comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

The CDP/debug-profile/login-reuse niche with the named `agent-browser` tool is highly distinct, but the trigger phrase "browser automation" is broad enough to overlap with general web-scraping or Playwright-style skills. That is minor overlap risk with closely related skills (anchor 4) rather than the minimal-conflict anchor 5.

4 / 5

Total

18

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

zenstory-ai/oh-story-claudecode

Path

skills/browser-cdp/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.