CtrlK
BlogDocsLog inGet started
Tessl Logo

build-review-interface

Build a custom browser-based annotation interface tailored to your data for reviewing LLM traces and collecting structured feedback. Use when you need to build an annotation tool, review traces, or collect human labels.

56

Quality

62%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/build-review-interface/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%

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

This skill provides a solid conceptual blueprint for building an annotation interface with good coverage of UI design, feedback collection, navigation, and testing. Its main weakness is the lack of executable code examples—for a skill about building an HTML-based tool, starter code or templates would dramatically improve actionability. The workflow could also benefit from explicit build sequencing and validation checkpoints.

Suggestions

Add a starter HTML/JS code template showing the basic scaffold with trace loading, Pass/Fail buttons, and auto-save functionality to make the skill copy-paste actionable.

Add an example data schema (e.g., a sample JSON trace format) so Claude knows what input structure to expect and can build the loader accordingly.

Restructure the build process into an explicit numbered workflow (Step 1: scaffold, Step 2: data loading, Step 3: annotation controls, Step 4: test) with validation checkpoints between steps.

Consider splitting the detailed Data Display guidelines and Testing section into separate referenced files to improve progressive disclosure.

DimensionReasoningScore

Conciseness

The skill is mostly efficient and avoids explaining concepts Claude already knows (no 'what is an annotation interface' preamble). However, some sections like the Data Display bullet list are somewhat verbose and could be tightened—several bullets convey overlapping ideas (collapse, toggle, hierarchy). The design checklist partially repeats guidance already given in the body.

2 / 3

Actionability

The skill provides concrete guidance on what to build (specific UI elements, keyboard shortcuts, testing steps) but lacks executable code examples. There are no HTML/CSS/JS snippets, no example data schema, and no concrete file structure. The Playwright testing section describes what to test but doesn't provide executable test code. For a 'build an HTML page' skill, at least a starter template or code skeleton would significantly improve actionability.

2 / 3

Workflow Clarity

The skill has a logical structure progressing from data display → feedback → navigation → testing, and the testing section has a clear numbered sequence. However, the overall build workflow lacks explicit sequencing—there's no clear 'Step 1: Create the HTML scaffold, Step 2: Add data loading, Step 3: Add annotation controls' progression. The testing section provides good validation steps but there's no feedback loop for fixing issues found during visual review or functional testing.

2 / 3

Progressive Disclosure

The content is well-organized with clear section headers and a logical flow from overview to details. However, for a skill of this length (~100 lines of substantive content), some sections like the detailed Data Display guidelines or the Testing section could be split into referenced files. There are no bundle files and no references to external files, making this somewhat monolithic for the breadth of topics covered.

2 / 3

Total

8

/

12

Passed

Description

75%

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 well-structured with a clear 'Use when' clause and a distinct niche around LLM trace annotation. Its main weakness is moderate specificity—it could enumerate more concrete actions the skill enables—and the trigger terms could cover more natural variations of how users might request this functionality.

Suggestions

Add more specific concrete actions like 'create label schemas, configure annotation workflows, export labeled datasets, set up multi-annotator review'

Expand trigger terms to include common variations like 'labeling tool', 'data labeling', 'RLHF feedback', 'human evaluation', 'evaluation UI'

DimensionReasoningScore

Specificity

The description names the domain (browser-based annotation interface) and some actions (reviewing LLM traces, collecting structured feedback), but doesn't list multiple concrete specific actions like creating label schemas, configuring annotation workflows, exporting labeled data, etc.

2 / 3

Completeness

Clearly answers both 'what' (build a custom browser-based annotation interface for reviewing LLM traces and collecting structured feedback) and 'when' (explicit 'Use when' clause with triggers: build an annotation tool, review traces, collect human labels).

3 / 3

Trigger Term Quality

Includes some relevant keywords like 'annotation tool', 'traces', 'human labels', and 'structured feedback', but misses common variations users might say such as 'labeling tool', 'RLHF', 'data labeling', 'evaluation UI', 'human evaluation', or 'review interface'.

2 / 3

Distinctiveness Conflict Risk

The combination of 'annotation interface', 'LLM traces', and 'structured feedback/human labels' creates a clear niche that is unlikely to conflict with general web development or data processing skills.

3 / 3

Total

10

/

12

Passed

Validation

100%

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

Validation11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository
hamelsmu/evals-skills
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.