CtrlK
BlogDocsLog inGet started
Tessl Logo

go-agent-development

编写 BK-CI Go Agent 代码时使用,例如 Agent API 调用、任务处理、并发模式、错误处理、日志记录和宿主侧工具开发。当用户要写 Go 构建机侧代码而不是后端 Kotlin 服务时优先使用。

60

Quality

76%

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 ./ai/skills/go-agent-development/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%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.

A concise, well-organized routing/habits skill whose structure and scope boundaries are clear, but whose guidance is principle-level rather than actionable: it tells Claude what to care about (goroutine lifecycle, error recovery, log quality) without any concrete Go patterns, examples, or verification steps to execute against.

Suggestions

Add concrete Go patterns for the named pitfalls — e.g., an errgroup/context-cancellation snippet showing goroutine exit paths, and error wrapping with %w plus structured log fields — so the guidance is executable rather than advisory.

Turn 快速指导 into a sequenced workflow with explicit checkpoints, e.g., 'before finishing a change, verify: every started goroutine has a shutdown path; every network call has retry/timeout; logs carry enough context to locate the failure'.

Remove redundancy by merging 快速指导 items 4–5 into the 延伸阅读 section and trimming 适用场景 items that restate the frontmatter description.

DimensionReasoningScore

Conciseness

The body is lean with no explanation of concepts Claude already knows, and every section is short. Minor redundancy keeps it below a 5: 适用场景 restates the frontmatter description, 快速指导 items 4–5 duplicate the 延伸阅读 section, and 高信号规则 partially overlaps 关键陷阱.

4 / 5

Actionability

Guidance stays at the level of high-level hints ("错误处理、日志、重试、清理和并发边界要先想清楚", "goroutine 启动容易,退出和异常恢复没管") with no concrete patterns, code, commands, or specific library usage for a coding skill. It is not entirely abstract (the link-out directives in items 4–5 are concrete actions), so it sits above a 1, but it is below a 3 because the core Go-side guidance lacks the specific steps to execute.

2 / 5

Workflow Clarity

快速指导 is a numbered 5-item list, giving a rough sequence (stability first, view lifecycles together, then check linked skills), but the items read as unordered advice rather than a sequenced workflow, and there are no validation checkpoints. No destructive/batch operations are involved, so no cap applies; it does not reach a 4's clear sequence with checkpoints.

3 / 5

Progressive Disclosure

The skill is under 50 lines, has no bundle files, and nothing inlined belongs in a separate file — per the rubric's simple-skill guideline this qualifies for the top score. Sections are clearly headed (适用场景/不适用场景/快速指导/高信号规则/关键陷阱/延伸阅读), and the cross-references to agent-module-architecture and dispatch-module-architecture are one level deep and clearly signaled.

5 / 5

Total

14

/

20

Passed

Description

83%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.

A strong description: concrete capability enumeration, explicit 'use when' guidance with a disambiguating condition, and natural trigger terms. The only refinement space is stating capabilities as actions rather than domains and covering a few more user-side synonyms.

DimensionReasoningScore

Specificity

The description enumerates six concrete capability areas ("Agent API 调用、任务处理、并发模式、错误处理、日志记录和宿主侧工具开发"), giving several specific actions with only minor gaps. It falls short of a 5 because these are stated as work domains ('并发模式', '错误处理') rather than the fully concrete actions of the top anchor, but is clearly above a 3's '1-2 concrete actions'.

4 / 5

Completeness

It explicitly answers both what ("Agent API 调用、任务处理、并发模式、错误处理、日志记录和宿主侧工具开发") and when with concrete trigger phrasing ("编写 BK-CI Go Agent 代码时使用…当用户要写 Go 构建机侧代码而不是后端 Kotlin 服务时优先使用"). The 'use when' clause is explicit and includes a concrete disambiguating condition, matching the top anchor; a 4 would require the 'when' to be less specific than it is.

5 / 5

Trigger Term Quality

Natural terms a user would say are present — "Go Agent", "BK-CI", "构建机侧代码", "Kotlin", "API 调用" — with good keyword coverage across the domain. A 5 would require synonyms/extensions users might also say (e.g., '构建机', 'devops agent' variants), which are only partially covered; it is well above the generic keyword level of a 3.

4 / 5

Distinctiveness Conflict Risk

It carves a clear niche (BK-CI Go Agent code) and explicitly steers away from backend Kotlin services, making mis-triggering unlikely. Minor overlap risk remains with sibling skills in the same domain (agent-module-architecture covers the same Agent from an architecture angle), keeping it below the minimal-conflict top anchor.

4 / 5

Total

17

/

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
TencentBlueKing/bk-ci
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.