CtrlK
BlogDocsLog inGet started
Tessl Logo

process-module-architecture

处理 BK-CI Process 流水线核心模块时使用,例如流水线 CRUD、构建启动、运行时状态、引擎控制、事件驱动和数据库持久化。当用户要改流水线主链路而不是外围支撑模块时优先使用。

61

Quality

77%

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/process-module-architecture/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

60%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.

The body is a well-organized, appropriately concise hub-and-spoke overview with good scoping and cross-skill routing. Its critical defect is that all five reference documents it routes to are missing from the bundle, leaving the skill a non-functional index, and beyond the file pointers it offers few concrete code-level anchors for the Process module.

Suggestions

Ship the five referenced documents (reference/1-api-layer.md through reference/5-event-driven.md) with the skill — or fix the paths to the actual files — since the entire navigation structure is built on them and none resolve in the current bundle.

Add a few concrete source anchors (e.g., key packages, entry-point classes, or file paths in the Process module such as the engine control or event-driven classes) so the guidance remains actionable even before opening the reference files.

Deduplicate the reference list: the same five file pointers appear both in 快速指导 item 3 and again in 延伸阅读; keep one canonical list to save tokens and avoid drift.

DimensionReasoningScore

Conciseness

The ~55-line body is lean and assumes domain competence — no generic concept explanations — with useful sections (适用/不适用场景, 高信号规则, 关键陷阱). It loses a point for redundancy: the identical five-file reference list appears twice (快速指导 item 3 and 延伸阅读), which could be collapsed. It is efficient with minor trimmable material (anchor 4), not the every-token-earns-its-place case of anchor 5.

4 / 5

Actionability

Guidance is partially concrete — a four-layer decomposition (入口层/服务层/引擎层/数据层) and named per-layer documents — but the referenced files (reference/1-api-layer.md … reference/5-event-driven.md) do not exist in the bundle, and the body gives no concrete source anchors (packages, classes, or file paths in the Process module) to act on without them. This matches anchor 3 ('some concrete guidance but incomplete; missing key details') rather than anchor 4's mostly-executable guidance.

3 / 5

Workflow Clarity

There is a coherent decision procedure: scope check via 适用/不适用场景, layer decomposition, per-topic reference routing, and cross-skill routing rules (items 4–6). This is an understanding/orientation skill with no destructive or batch operations, so validation checkpoints are not required; sequence is clear with only minor gaps (e.g., no guidance on what to do if a referenced doc is missing), matching anchor 4.

4 / 5

Progressive Disclosure

The in-body structure is well signaled and one level deep, but every referenced path — "reference/1-api-layer.md" through "reference/5-event-driven.md" — is absent: the bundle contains no reference/, references/, scripts/, or assets/ directories at all, and the paths use a 'reference/' spelling that does not match the conventional 'references/' layout. The navigation the entire skill depends on does not resolve, which is a broken-disclosure defect below the midpoint (anchor 3's 'references present but not clearly signaled' understates it; anchor 2's structurally broken organization is the closer fit).

2 / 5

Total

13

/

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: it names the domain, lists six concrete capability areas, and gives an explicit positive-and-negative trigger clause that disambiguates it from peripheral BK-CI modules. Weaknesses are minor — a generic leading verb and a few missing natural trigger synonyms.

DimensionReasoningScore

Specificity

The description enumerates several concrete capability areas — "流水线 CRUD、构建启动、运行时状态、引擎控制、事件驱动和数据库持久化" — giving specific coverage of the module's scope, though the only verb is the generic "处理", so the actions read as noun-phrase capability areas rather than full verb-led actions. This sits above anchor 3 (only 1–2 concrete actions) but the generic verb keeps it below anchor 5's fully action-verb comprehensive list.

4 / 5

Completeness

Both parts are explicit: the 'what' is "处理 BK-CI Process 流水线核心模块…例如流水线 CRUD、构建启动…" and the 'when' is the concrete trigger clause "当用户要改流水线主链路而不是外围支撑模块时优先使用", which even includes a negative boundary. It clearly matches anchor 5 and is not merely the weaker 'when could be more explicit' case of anchor 4.

5 / 5

Trigger Term Quality

Natural terms a BK-CI user would say are present: "流水线", "构建", "Process", "流水线主链路". Coverage is good but misses common variations such as the English word "pipeline", "编排", or file/module-name triggers, so it falls between anchor 4 (good coverage, a few natural terms missing) and anchor 5, closer to 4.

4 / 5

Distinctiveness Conflict Risk

The niche is well carved out — BK-CI's Process main chain, explicitly contrasted against "外围支撑模块" — and it names sibling-module territory to avoid. Minor overlap remains with closely related dispatch/agent/worker skills around build execution, so it is 'mostly distinct with minor overlap risk' (anchor 4) rather than fully minimal-conflict (anchor 5).

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.