replayio

Use when you need to record or inspect an agent browser run in Replay, test a local app with the host agent browser using Replay Chromium, or use the Replay MCP server for deeper debugging of an uploaded recording.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./plugins/replayio/skills/replayio/SKILL.md

Quality

Content

77%Weight 40%Scale 1-3

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

The content is highly actionable with concrete executable examples and a well-validated multi-step workflow. Its main weaknesses are repetition that inflates token cost and overlap between the inline body and the reference files that undercuts progressive disclosure.

Suggestions

Define the AGENT_BROWSER_EXECUTABLE_PATH / RECORD_* export block and the upload command once, then reference them by name elsewhere instead of repeating the full snippets.

Move the detailed JS interaction patterns and debugging snippets into the existing reference files, keeping SKILL.md as a concise overview that points to them.

Collapse the overlapping sections ('The Reliable Path', 'Reliable Agent Browser Workflow', 'Prerequisites') into a single canonical sequence to remove duplicated steps.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and actionable, but the AGENT_BROWSER_EXECUTABLE_PATH export block repeats roughly four times and 'replayio upload-all \|\| replayio upload' is restated multiple times, so it 'could be tightened' and includes some unnecessary duplication rather than earning a lean score of 3.	2 / 3
Actionability	It provides fully executable bash exports, replayio CLI commands, and copy-paste-ready JS browser-API snippets with specific env vars and locators, matching the 'fully executable code/commands; copy-paste ready' anchor.	3 / 3
Workflow Clarity	'The Reliable Path' is a 10-step numbered sequence with explicit validation checkpoints (test -x, curl -I, replayio info/whoami, verify reachability) plus a Close-When-Done contract executed before reporting results, satisfying the 'clear sequence with explicit validation steps' anchor.	3 / 3
Progressive Disclosure	References to references/cli.md and references/workflows.md are real, one level deep, and signaled in a References section, but the body carries large inline detail that duplicates the reference files (same env blocks and JS patterns), fitting 'content that should be separate is inline' rather than a clean overview.	2 / 3
	Total	10 / 12 Passed

Description

77%Weight 40%Scale 1-3

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

The description is specific, complete, and trigger-driven, clearly answering both what the skill does and when to use it. Its main weakness is jargon-heavy trigger phrasing and a somewhat generic leading trigger that slightly raises conflict risk.

Suggestions

Lead with the most user-natural trigger (e.g., 'Use when the user asks to record or debug a browser session with Replay') before the agent-browser/MCP variants.

Add common keyword variations such as 'Replay recording', 'browser session recording', and 'time-travel debugging' to broaden natural trigger coverage.

Tighten jargon like 'host agent browser' to plain terms where possible to reduce overlap with generic browser skills.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — 'record or inspect an agent browser run in Replay', 'test a local app with the host agent browser using Replay Chromium', and 'use the Replay MCP server for deeper debugging of an uploaded recording' — matching the anchor for listing several specific actions.	3 / 3
Completeness	An explicit 'Use when you need to...' clause states the trigger conditions, and the body of the sentence clearly states what the skill does (record/inspect/test/debug with Replay/agent browser/MCP), answering both what and when.	3 / 3
Trigger Term Quality	It includes some natural terms ('record', 'Replay', 'test a local app', 'debugging') but leans on internal jargon ('agent browser run', 'host agent browser', 'Replay MCP server') and misses common variations like 'Replay recording' or 'browser session'; this fits 'some relevant keywords but missing common variations' rather than full coverage.	2 / 3
Distinctiveness Conflict Risk	The Replay-specific niche is fairly distinct, but the leading trigger 'agent browser run' is somewhat generic and could overlap with other browser skills; it is not yet a 'clear niche with distinct triggers' free of overlap.	2 / 3
	Total	10 / 12 Passed

Validation

100%

Warnings & errors only

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: openai/plugins
Path: plugins/replayio/skills/replayio/SKILL.md
Commit: 11c74d6

Reviewed: 17 days ago

Table of Contents

Discovery Implementation Validation

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.