llm-monitoring-dashboard

Auto-generates an LLM usage monitoring page in a PM admin dashboard. Tokuin CLI-based token/cost/latency tracking + user ranking system + inactive user tracking + data-driven PM insights + Cmd+K global search + per-user drilldown navigation. Supports OpenAI/Anthropic/Gemini/OpenRouter.

Install with Tessl CLI

npx tessl i github:supercent-io/skills-template --skill llm-monitoring-dashboard

What are skills?

1.86x

Quality

52%

Does it follow best practices?

Impact

99%

1.86x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./.agent-skills/llm-monitoring-dashboard/SKILL.md

SKILL.md

Review

Evals

Discovery

50%

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 excels at specificity and distinctiveness, providing detailed feature lists and a clear niche for LLM usage monitoring dashboards. However, it critically lacks any 'Use when...' guidance, making it difficult for Claude to know when to select this skill. The technical jargon ('Tokuin CLI-based', 'PM admin dashboard') may not match natural user language.

Suggestions

Add a 'Use when...' clause with trigger terms like 'track API usage', 'monitor LLM costs', 'usage dashboard', 'API spending', 'token consumption'

Replace or supplement technical terms like 'Tokuin CLI-based' with user-friendly language such as 'command-line tool for tracking' or explain what Tokuin is

Include common user phrases like 'how much am I spending on AI', 'track OpenAI costs', 'monitor API usage across teams'

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions: 'token/cost/latency tracking', 'user ranking system', 'inactive user tracking', 'data-driven PM insights', 'Cmd+K global search', 'per-user drilldown navigation'. Very detailed about what it builds.	3 / 3
Completeness	Describes WHAT it does comprehensively but completely lacks a 'Use when...' clause or any explicit trigger guidance. Per rubric guidelines, missing explicit trigger guidance should cap completeness at 2, and this has no 'when' component at all.	1 / 3
Trigger Term Quality	Includes some relevant terms like 'LLM usage monitoring', 'token', 'cost', 'latency', 'OpenAI/Anthropic/Gemini/OpenRouter', but uses technical jargon ('Tokuin CLI-based', 'PM admin dashboard') that users may not naturally say. Missing common variations like 'API costs', 'usage dashboard', 'spending tracker'.	2 / 3
Distinctiveness Conflict Risk	Very specific niche: LLM usage monitoring dashboards with specific providers (OpenAI/Anthropic/Gemini/OpenRouter) and specific features (Tokuin CLI, PM admin context). Unlikely to conflict with other skills due to its narrow, well-defined scope.	3 / 3
	Total	9 / 12 Passed

Implementation

55%

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

This skill excels at actionability with fully executable code and clear workflow sequencing with safety checkpoints. However, it severely violates conciseness principles by embedding enormous code blocks inline (particularly the 500+ line HTML dashboard) and lacks progressive disclosure—everything is in one massive file rather than appropriately split across referenced documents.

Suggestions

Extract the large HTML dashboard (Step 3 Option B) into a separate referenced file like `dashboard-template.html` and link to it from SKILL.md

Move the Next.js component code blocks to separate files (e.g., `NEXTJS_SETUP.md`) and reference them with clear navigation links

Remove explanatory text that Claude already knows (e.g., what environment variables are, how cron works, what JSONL format is)

Consolidate the CSS design tokens and reusable bash functions into referenced utility files rather than repeating inline

Dimension	Reasoning	Score
Conciseness	The skill is extremely verbose at ~1000+ lines, with extensive inline code that could be referenced externally. It includes unnecessary explanations (e.g., explaining what environment variables are, what cron does) and massive HTML/CSS/JS blocks that bloat the document significantly.	1 / 3
Actionability	The skill provides fully executable, copy-paste ready code throughout—complete bash scripts, Python modules, TypeScript files, and a full HTML dashboard. Every step has concrete, runnable commands with expected outputs documented.	3 / 3
Workflow Clarity	The workflow is clearly sequenced (Steps 0-6) with explicit validation checkpoints. Step 0 is a mandatory safety gate that halts on failures, and the skill includes feedback loops (validate → fix → retry) for risky operations like live API calls.	3 / 3
Progressive Disclosure	The skill is a monolithic wall of text with no external file references for detailed content. Massive code blocks (500+ lines of HTML/JS) are inline rather than in separate files. The References section exists but the core content lacks appropriate splitting.	1 / 3
	Total	8 / 12 Passed

Validation

81%

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 — 9 / 11 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1381 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning

	Total	9 / 11 Passed

Reviewed: about 19 hours 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.