modal
Cloud computing platform for running Python on GPUs and serverless infrastructure. Use when deploying AI/ML models, running GPU-accelerated workloads, serving web endpoints, scheduling batch jobs, or scaling Python code to the cloud. Use this skill whenever the user mentions Modal, serverless GPU compute, deploying ML models to the cloud, serving inference endpoints, running batch processing in the cloud, or needs to scale Python workloads beyond their local machine. Also use when the user wants to run code on H100s, A100s, or other cloud GPUs, or needs to create a web API for a model.
70
86%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong skill description that clearly identifies the platform (Modal), lists concrete actions (deploying models, serving endpoints, scheduling jobs), and provides comprehensive trigger guidance with natural user terms. It uses proper third-person voice throughout and covers both common and specific use cases including named GPU hardware. The description is slightly verbose with some redundancy between the two 'Use when' clauses, but this doesn't detract from its effectiveness.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: deploying AI/ML models, running GPU-accelerated workloads, serving web endpoints, scheduling batch jobs, scaling Python code to the cloud, creating web APIs for models. | 3 / 3 |
Completeness | Clearly answers both 'what' (cloud computing platform for running Python on GPUs and serverless infrastructure) and 'when' with explicit 'Use when...' and 'Use this skill whenever...' clauses listing detailed trigger scenarios. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: Modal, serverless GPU, H100s, A100s, cloud GPUs, ML models, inference endpoints, batch processing, web API, Python workloads. These are terms users would naturally use when needing this skill. | 3 / 3 |
Distinctiveness Conflict Risk | Clearly targets a specific niche — Modal as a cloud computing platform for GPU workloads and serverless Python. The mention of specific GPU types (H100s, A100s), Modal by name, and serverless GPU compute makes it highly distinguishable from generic cloud or Python skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
72%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, well-organized skill that covers Modal comprehensively with executable code examples and excellent progressive disclosure to reference files. Its main weakness is the lack of validation checkpoints and error recovery guidance in workflows (e.g., what to do when deployment fails, how to verify GPU allocation). There's also some minor verbosity in the overview and 'When to Use' sections that could be trimmed.
Suggestions
Add validation/verification steps to workflow patterns (e.g., 'Verify deployment: modal app list | grep my-app', 'Check logs: modal app logs <name>' after deploy steps)
Remove or significantly trim the 'When to Use This Skill' section since it largely duplicates the skill description in the frontmatter and is guidance for skill selection, not for executing tasks
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is generally well-structured but includes some unnecessary content like the 'When to Use This Skill' section (which duplicates the frontmatter description) and minor explanatory padding (e.g., 'Everything in Modal is defined as code — no YAML, no Dockerfiles required'). The overview bullet list and some descriptions could be tighter, but most content earns its place with concrete code examples. | 2 / 3 |
Actionability | The skill provides fully executable, copy-paste-ready code examples for every major feature: app creation, image building, GPU usage, volumes, secrets, web endpoints, scheduling, scaling, and complete workflow patterns. CLI commands are concrete and specific with a comprehensive reference table. | 3 / 3 |
Workflow Clarity | The workflow patterns section provides good complete examples, and the authentication section has a clear 3-step fallback sequence. However, there are no explicit validation checkpoints or error recovery steps in the multi-step workflows (e.g., no 'verify deployment succeeded' step, no 'check volume write completed' validation, no error handling guidance for failed GPU allocation or container builds). | 2 / 3 |
Progressive Disclosure | Excellent progressive disclosure structure: the SKILL.md serves as a clear overview with concise examples for each concept, and every section ends with a well-signaled one-level-deep reference to detailed documentation. The 'Reference Files' section at the end provides a clean navigation index to all 12 reference files with clear descriptions. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
Total | 10 / 11 Passed | |
- Repository
- K-Dense-AI/claude-scientific-skills
- Commit
cbcae7b
- 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.