CtrlK
BlogDocsLog inGet started
Tessl Logo

llm-runtime-architecture

Use when designing how one canonical agent core runs across Codex, Claude Code, Gemini CLI, Cursor, and AGENTS.md-compatible tools.

58

Quality

66%

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 ./.agents/skills/llm-runtime-architecture/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

71%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.

The content is exceptionally concise and well-structured with a clear output schema, but it lacks executable commands and an internal validation checkpoint, which cap actionability and workflow clarity at the midpoint.

Suggestions

Add a concrete verification command (e.g. how to check `.agentlas/global-commands.json` parses) as an explicit step so the workflow includes a validation checkpoint.

Provide a small executable snippet or a filled-in row of the runtime matrix as an example to lift actionability toward fully executable guidance.

Show one example adapter skeleton or template so 'keep adapters thin' becomes concrete rather than abstract.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence: a numbered procedure and an output schema with no padded explanation of what runtimes or adapters are; every line earns its place, matching the 'lean and efficient' anchor.

5 / 5

Actionability

The procedure gives concrete high-level directives (name entry point, global command, adapter files, tools, memory access, limitations, verification) and a precise output schema, but there is no executable code or commands and key details for executing each step are absent, fitting the 'some concrete guidance but incomplete' anchor.

3 / 5

Workflow Clarity

Five clearly sequenced steps are listed, but the operation involves creating/packaging an agent and writing files (a build/batch-style action) with no validation checkpoint or verification step inside the workflow; the guideline caps workflow clarity at 3 for such operations lacking validation.

3 / 5

Progressive Disclosure

A short skill with well-organized Procedure and Output sections, no external references needed, and no nested or buried references; under the simple-skill exception this matches the 'well-organized sections' anchor for a 5.

5 / 5

Total

16

/

20

Passed

Description

62%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.

The description has an explicit 'Use when' trigger and names concrete runtimes, giving it solid completeness and distinctiveness, but the single high-level action and sparse natural trigger phrases limit specificity and trigger-term quality.

Suggestions

Add more concrete actions beyond 'designing' (e.g. 'name entry points and adapters, document each runtime's tools and limits') to lift specificity.

Expand the trigger with natural user phrasings like 'porting an agent across runtimes', 'multi-runtime agent', or 'AGENTS.md-compatible tools' for better trigger-term coverage.

Add file extensions or canonical artifact names (e.g. `.agentlas/global-commands.json`) to make triggers more distinctive.

DimensionReasoningScore

Specificity

The description names the domain (canonical agent core running across multiple runtimes) and one concrete action ('designing how one canonical agent core runs across' named tools), but the action is a single high-level activity rather than several specific concrete actions; closest to the 'names domain and 1-2 concrete actions' anchor and not the 'several specific actions' anchor above.

3 / 5

Completeness

It has both a 'what' (designing how a canonical agent core runs across runtimes) and an explicit 'when' ('Use when designing...'), but the 'when' is a single narrow scenario rather than a range of concrete trigger phrases; matches the 'both what and when, when could be more explicit' anchor.

4 / 5

Trigger Term Quality

It includes relevant runtime names (Codex, Claude Code, Gemini CLI, Cursor, AGENTS.md) and the trigger 'Use when designing', but it is missing the common natural phrasings a user would actually say (e.g. 'agent runs on', 'port agent to', 'multi-runtime') and synonyms; partial coverage per the anchor.

3 / 5

Distinctiveness Conflict Risk

The named runtimes and 'canonical agent core across runtimes' framing carve a fairly distinct niche with minimal overlap risk against generic skills; minor overlap risk with sibling agent-authoring skills keeps it just below the 'clear niche' anchor.

4 / 5

Total

14

/

20

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
agentlas-ai/Agentlas-OS
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.