roboflow-inference
Deployment option comparison (serverless, dedicated, self-hosted, batch) and Workflow execution patterns. For raw API URL patterns, auth, and request/response formats, see roboflow-api-reference.
53
58%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/inference/SKILL.mdQuality
Discovery
32%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 provides a reasonable topic overview by listing deployment types and mentioning workflow execution patterns, and the cross-reference to the API reference skill is a nice touch for disambiguation. However, it lacks an explicit 'Use when...' clause, doesn't describe concrete actions the skill enables, and misses natural trigger terms a user would employ when needing deployment guidance.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user asks which Roboflow deployment option to choose, how to deploy a model, or needs help with inference workflows.'
List concrete actions the skill performs, e.g., 'Compares serverless vs. dedicated vs. self-hosted vs. batch deployment tradeoffs, explains workflow execution patterns, and recommends deployment configurations.'
Include natural trigger terms users would say, such as 'deploy model', 'hosting options', 'inference speed', 'Roboflow deployment', 'run predictions', or 'which deployment type'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (deployment options, workflow execution) and lists specific deployment types (serverless, dedicated, self-hosted, batch), but doesn't describe concrete actions beyond 'comparison' and 'patterns' — these are somewhat vague about what the skill actually does. | 2 / 3 |
Completeness | Describes 'what' at a high level (deployment comparison and workflow patterns) but has no explicit 'Use when...' clause or equivalent trigger guidance. The cross-reference to roboflow-api-reference is helpful for disambiguation but doesn't substitute for when-to-use guidance. Per rubric, missing 'Use when' caps completeness at 2, and the 'what' is also weak, so this scores 1. | 1 / 3 |
Trigger Term Quality | Includes some relevant keywords like 'serverless', 'dedicated', 'self-hosted', 'batch', 'deployment', and 'workflow execution patterns', but misses natural user phrases like 'deploy model', 'run inference', 'which deployment should I use', or 'Roboflow hosting options'. | 2 / 3 |
Distinctiveness Conflict Risk | The explicit mention of deployment option types and the cross-reference to roboflow-api-reference help distinguish it from a generic API skill, but 'workflow execution patterns' is broad and could overlap with other workflow-related skills. The disambiguation note is a positive signal but the scope is still somewhat ambiguous. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured deployment and inference skill that serves effectively as a hub document, providing clear decision frameworks (deployment options table, workflow authoring modes, video options) with actionable CLI examples and appropriate delegation to companion files. Its main weakness is moderate verbosity — some sections repeat guidance (e.g., video options appear twice, workflow authoring caveats are restated) and the introductory callout blocks are lengthy. Overall it's a strong reference skill with good progressive disclosure.
Suggestions
Consolidate the video deployment options — they appear both in the callout block at the top and again under 'When to Use Which'; keep the detailed version in one place only.
Trim the two introductory blockquotes (plugin/MCP fallback and workflow authoring modes) which together consume ~30 lines before the skill's core content begins; move the authoring modes detail entirely to workflows.md and keep only a one-line pointer.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly comprehensive but includes some unnecessary verbosity — e.g., the lengthy preamble about plugin vs MCP fallback, repeated cross-references to the same workflows resource, and explanatory prose like 'Problem it solves' and 'Pick it when' sections that could be condensed into the comparison table. However, most content is informational rather than explaining things Claude already knows. | 2 / 3 |
Actionability | The skill provides concrete CLI commands for batch processing (end-to-end example with staging, submitting, monitoring, exporting), specific MCP tool names with clear purposes, deployment option specifics (subdomains, upload limits), and executable bash examples. The guidance is copy-paste ready and specific. | 3 / 3 |
Workflow Clarity | The batch processing flow is clearly sequenced (1-5 steps) with explicit monitoring/polling step. The workflow authoring section clearly distinguishes Mode A vs Mode B with decision criteria. The video deployment section presents three options with clear differentiation. Decision trees for 'When to Use Which' are well-structured. | 3 / 3 |
Progressive Disclosure | Content is well-structured as an overview with clear one-level-deep references to companion files: workflows.md, workflow-templates.md, local-tooling.md, batch-staging.md, batch-jobs.md, and MCP resource URIs. The main skill covers the overview and decision-making while delegating details appropriately. | 3 / 3 |
Total | 11 / 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.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- roboflow/computer-vision-skills
- Commit
02936d5
- 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.