peekaboo
Capture and automate macOS UI with the Peekaboo CLI.
56
43%
Does it follow best practices?
Impact
77%
1.92xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./openclaw/skills/peekaboo/SKILL.mdQuality
Discovery
22%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 too terse and vague, failing to enumerate specific capabilities of the Peekaboo CLI or provide any guidance on when Claude should select this skill. While it names the tool and platform (macOS), it lacks the concrete actions and trigger terms needed for effective skill selection among many options.
Suggestions
Add a 'Use when...' clause specifying trigger scenarios, e.g., 'Use when the user needs to take screenshots, inspect UI elements, automate clicks, or interact with macOS application interfaces.'
List specific concrete actions the Peekaboo CLI supports, such as 'take screenshots, read accessibility trees, click UI elements, list open windows, inspect element properties.'
Include natural trigger terms users might say, such as 'screenshot', 'screen capture', 'click button', 'UI element', 'window list', 'GUI automation', 'accessibility tree'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description says 'Capture and automate macOS UI' which is vague. It doesn't list specific concrete actions like taking screenshots, clicking elements, reading accessibility trees, or other specific capabilities. 'Capture' and 'automate' are broad terms. | 1 / 3 |
Completeness | The description only vaguely addresses 'what' (capture and automate macOS UI) and completely lacks a 'when' clause. There is no 'Use when...' or equivalent trigger guidance, which per the rubric should cap completeness at 2, but the 'what' is also weak, so this scores a 1. | 1 / 3 |
Trigger Term Quality | Includes 'macOS UI', 'Peekaboo CLI', and 'capture'/'automate' which are somewhat relevant, but misses natural user terms like 'screenshot', 'click button', 'UI testing', 'accessibility', 'screen capture', or 'GUI automation'. | 2 / 3 |
Distinctiveness Conflict Risk | Mentioning 'Peekaboo CLI' and 'macOS UI' provides some distinctiveness, but 'automate macOS UI' could overlap with other macOS automation skills (e.g., AppleScript, Automator, or other UI automation tools). The niche is somewhat defined but not sharply delineated. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill with excellent concrete examples and good coverage of the Peekaboo CLI's capabilities. Its main weaknesses are the inlined exhaustive command catalog (which inflates token usage without adding proportional value) and the lack of validation/error-recovery steps in workflows, which is important for inherently fragile UI automation tasks.
Suggestions
Move the exhaustive feature/command catalog to a separate REFERENCE.md file and keep SKILL.md focused on quickstart, common patterns, and key parameters.
Add validation checkpoints to workflows — e.g., after `peekaboo see`, verify elements were found before clicking; after `peekaboo permissions`, check output before proceeding.
Add a brief troubleshooting or error-recovery section for common failures (permissions denied, element not found, window not focused).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly efficient and avoids explaining basic concepts, but the exhaustive feature listing (Core/Interaction/System/Vision sections) is essentially a command catalog that could be referenced externally rather than inline. Some redundancy exists between the feature list and the examples. | 2 / 3 |
Actionability | Excellent concrete, copy-paste-ready bash commands throughout. Every example is fully executable with specific flags, arguments, and realistic values. The quickstart and detailed examples cover the most common workflows clearly. | 3 / 3 |
Workflow Clarity | The 'See -> click -> type' flow is well-sequenced and represents the most reliable pattern. However, there are no explicit validation checkpoints or error recovery steps — e.g., what to do if permissions fail, if `see` returns no elements, or if a click misses its target. For UI automation (which can be fragile), feedback loops would be valuable. | 2 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and headers, but the full feature catalog is inlined rather than referenced externally. The skill would benefit from moving the exhaustive command list to a separate reference file and keeping SKILL.md focused on the quickstart, common patterns, and key parameters. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 8 / 11 Passed | |
- Repository
- trpc-group/trpc-agent-go
- Commit
0c1ec2b
- 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.