azure-hosted-copilot-sdk
Build, deploy, modify GitHub Copilot SDK apps on Azure. MANDATORY when codebase contains @github/copilot-sdk or CopilotClient — use this skill instead of azure-prepare. PREFER OVER azure-prepare when codebase contains copilot-sdk markers. WHEN: copilot SDK, @github/copilot-sdk, copilot-powered app, deploy copilot app, add feature, modify copilot app, BYOM, bring your own model, CopilotClient, createSession, sendAndWait, azd init copilot. DO NOT USE FOR: general web apps without copilot SDK (use azure-prepare), Copilot Extensions, Foundry agents (use microsoft-foundry).
85
81%
Does it follow best practices?
Impact
Pending
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 an excellent skill description that clearly defines its scope, provides abundant natural trigger terms, and proactively addresses potential conflicts with related skills (azure-prepare, microsoft-foundry). The explicit MANDATORY, PREFER OVER, WHEN, and DO NOT USE FOR sections make it highly actionable for skill selection. The description is concise yet comprehensive.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions: 'Build, deploy, modify GitHub Copilot SDK apps on Azure.' Also specifies what it covers (BYOM, CopilotClient, createSession, sendAndWait, azd init copilot) and what it doesn't cover. | 3 / 3 |
Completeness | Clearly answers both 'what' (build, deploy, modify GitHub Copilot SDK apps on Azure) and 'when' (explicit WHEN clause with trigger terms, MANDATORY conditions based on codebase markers, and DO NOT USE FOR exclusions that clarify boundaries). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would say: 'copilot SDK', '@github/copilot-sdk', 'copilot-powered app', 'deploy copilot app', 'BYOM', 'bring your own model', 'CopilotClient', 'createSession', 'sendAndWait', 'azd init copilot'. These are highly specific terms a user or codebase would naturally contain. | 3 / 3 |
Distinctiveness Conflict Risk | Exceptionally distinctive with explicit conflict resolution: specifies when to use this over 'azure-prepare', and explicitly excludes 'Copilot Extensions' and 'Foundry agents' with redirects to other skills. The niche is very clearly defined around the GitHub Copilot SDK specifically. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
62%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 well-structured routing and orchestration framework for Copilot SDK apps on Azure, with clear workflow sequencing and good use of reference files for progressive disclosure. However, the detection/routing section is repetitively verbose (three warning callouts saying the same thing), and the skill relies heavily on reference files for actionable details while providing limited executable code in the main body itself.
Suggestions
Consolidate the three redundant ⚠️ warning callouts in the detection section into a single, prominent warning to reduce token waste.
Add at least one concrete, executable BYOM code snippet inline (e.g., the DefaultAzureCredential + bearerToken pattern) rather than deferring entirely to auth-best-practices.md, since this is called out as MANDATORY.
Remove the 'Generic prompts that MUST trigger this skill' table — this is routing metadata that Claude can infer from the detection procedure and doesn't need spelled out with examples.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The detection section is quite verbose with redundant tables and repeated warnings (three ⚠️ callouts saying essentially the same thing about routing). The routing table and steps 2A-4 are lean and efficient, but the mandatory first check section could be significantly tightened. | 2 / 3 |
Actionability | Provides concrete commands (azd init --template, docker info) and clear routing tables, but most steps delegate to reference files for actual implementation details. The skill itself lacks executable code examples for key operations like BYOM configuration, SDK integration patterns, or deployment commands beyond the initial scaffold. | 2 / 3 |
Workflow Clarity | The workflow is clearly sequenced: detection → route → scaffold/deploy → model config → deploy. Steps are numbered, routing is table-driven with clear conditions, and the deploy step explicitly sequences sub-skills (azure-prepare → azure-validate → azure-deploy). The detection procedure has explicit validation checkpoints. | 3 / 3 |
Progressive Disclosure | References to supporting files (references/copilot-sdk.md, references/deploy-existing.md, references/azure-model-config.md, references/auth-best-practices.md, references/existing-project-integration.md) are well-signaled and one level deep. However, no bundle files were provided, so we cannot verify these references exist. The main file's detection section is overly detailed inline content that could be more concise with a reference. | 2 / 3 |
Total | 9 / 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
- microsoft/github-copilot-for-azure
- Commit
771a666
- 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.