CtrlK
BlogDocsLog inGet started
Tessl Logo

auto-build-repair

Headless, bounded repair of custom-code build failures in an already-generated Azure SDK PR. Thin wrapper over the shared azure-sdk-mcp:azsdk_customized_code_update engine in custom-code-only scope (editScope: CustomCode); the skill owns the iterate-until-green loop, capped by a per-language `maxIterations` read from repair-config.yml. WHEN: Copilot cloud agent runs on a release-planner Auto SDK PR labeled `auto-sdk-build-fix` that fails to build because of custom (non-generated) code. DO NOT USE FOR: full TypeSpec migrations, spec edits, API design review, manual fixing. INVOKES: azure-sdk-mcp:azsdk_customized_code_update.

67

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

77%

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

A highly actionable, well-sequenced repair workflow with strong validation and stop conditions. It is held back by noticeable redundancy across sections and a lack of progressive disclosure into reference files.

Suggestions

Collapse the repeated editScope/omit-tspProjectPath/do-not-hand-edit guidance into the Call shape and Hard Rules sections only, removing the restatements in the intro, engine, and bounds sections to cut tokens.

Move the inline 'Requires azsdk >= 0.6.22' version note (and any other time-sensitive version pins) into a short 'Requirements' or 'Deprecated/old patterns' subsection so it does not penalize conciseness.

Offload the detailed Stop Conditions table and Scope matrix into a reference file under references/ (e.g. STOP_CONDITIONS.md) and link to it one level deep, which would also let the cited repair-config.yml be verified as part of the bundle.

DimensionReasoningScore

Conciseness

Operational detail is domain-specific rather than beginner-concept bloat, but the same constraints (editScope: CustomCode, omit tspProjectPath, do not hand-edit, do not touch spec inputs) are restated 4-5 times across intro, engine, scope, call shape, bounds, and a Hard Rules recap, and the inline time-sensitive version note pads the token budget.

2 / 3

Actionability

Provides a concrete call shape with real parameters, a numbered workflow, and specific terminal error codes (SpecChangeRequired, RegenerateFailed) that map directly to actions.

3 / 3

Workflow Clarity

A clearly sequenced 0-5 workflow with an explicit feedback loop (re-invoke while the error set shrinks), validation via BuildResult inspection, and enumerated stop conditions.

3 / 3

Progressive Disclosure

Sections are well organized, but the skill is a >50-line monolith with all detail inline, no bundle reference files, and the one cited co-located file (repair-config.yml) is not present in the bundle.

2 / 3

Total

10

/

12

Passed

Description

85%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A dense, specific description that answers what, when, and what-not-to-use-for with a clear niche and third-person voice. The main weakness is trigger terms that are label/technical-jargon leaning rather than natural user phrasing.

DimensionReasoningScore

Specificity

Lists multiple concrete actions: 'Headless, bounded repair of custom-code build failures', 'Thin wrapper over the shared azure-sdk-mcp:azsdk_customized_code_update engine', 'owns the iterate-until-green loop, capped by a per-language maxIterations'.

3 / 3

Completeness

Clearly answers both what (bounded custom-code repair via a named engine) and when via an explicit 'WHEN:' clause plus a 'DO NOT USE FOR:' boundary.

3 / 3

Trigger Term Quality

Relevant triggers are present ('auto-sdk-build-fix', 'custom-code build failures', 'Auto SDK PR') but lean toward label/technical jargon rather than the natural phrasings a user would say, and miss common variations.

2 / 3

Distinctiveness Conflict Risk

A clear niche (custom-code-only repair of an already-generated Azure SDK PR via one shared engine) with a DO NOT USE FOR list distinguishing it from migrations, spec edits, and manual fixing.

3 / 3

Total

11

/

12

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository
Azure/azure-sdk-for-net
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.