CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

spec-workflow

Use when medium-to-large changes need explicit requirements, technical design, and task planning before implementation, especially for multi-module work, unclear acceptance criteria, or architecture-heavy requests.

Quality

82%

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

The canonical home for this skill is spec-workflow in TencentCloudBase/CloudBase-AI-Toolkit

SKILL.md
Quality
Evals
Security

Quality

Content

73%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 solid, well-sequenced planning workflow with copy-paste-ready templates for requirements (EARS) and tasks, and explicit confirmation gates between phases. The main weaknesses are the triple-stated use/skip decision guidance (token waste) and a Design phase that is less concretely scaffolded than the others.

Suggestions

Consolidate the three overlapping decision sections ('Activation Contract', 'When to use this skill', 'Decision rule') into a single use/skip section — the repeated trigger guidance is the largest source of wasted tokens.

Add a short design.md template or annotated example (parallel to the EARS pattern and task format) so the Design phase is as actionable as Requirements and Tasks.

Fold the 'Common mistakes / gotchas' entries into the phase guidance they relate to, trimming items that merely restate the confirmation rule stated elsewhere.

DimensionReasoningScore

Conciseness

The use/skip decision guidance is stated three separate times ('Activation Contract' with 'Use this first when'/'Do NOT use for', 'When to use this skill', and 'Decision rule' with 'Use the full workflow when'/'Skip the full workflow when'), which is noticeable redundancy; the rest of the body is tight bullets with no over-explanation of concepts Claude already knows. This fits anchor 3 (mostly efficient but could be tightened) rather than 4, where trimming would be only minor.

3 / 5

Actionability

Concrete scaffolds are provided: exact artifact paths ('specs/<spec_name>/requirements.md', 'design.md', 'tasks.md'), a copy-paste EARS pattern template with a worked example ('When the user submits the form, the booking system shall validate required fields...'), and a task-format template ('- [ ] 1. Task title ... _Requirement: 1'). Anchor 4 fits: mostly executable guidance with minor gaps — the Design phase ('Describe architecture and module boundaries', 'Explain technology choices and trade-offs') has no comparable template or example.

4 / 5

Workflow Clarity

The four phases (Requirements -> Design -> Tasks -> Execution) are explicitly sequenced with user-confirmation checkpoints between each: 'Get confirmation before moving to the next phase', 'Require confirmation between requirements, design, and task breakdown', and 'Only start implementation after the user confirms the task plan', plus an entry 'Minimal checklist'. This matches anchor 5 (clear sequence with explicit validation steps and a checklist); anchor 4 falls short because all inter-phase checkpoints are present, and no destructive/batch-operation cap applies to a planning workflow.

5 / 5

Progressive Disclosure

No bundle files exist, and the single-file body is well organized with clear sections; sibling-skill references ('../ui-design/SKILL.md', '../postgresql-development-cloudbase/SKILL.md') are clearly signaled and one level deep. Anchor 4 fits (good structure, minor organization gaps) rather than 5, since the redundant decision sections and 'Common mistakes' list leave minor organization gaps and there is no overview/reference split to evaluate.

4 / 5

Total

16

/

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 that explicitly answers both what the skill does and when to use it, with a proper 'Use when...' clause and multiple concrete trigger conditions. Minor room for improvement lies in adding common trigger synonyms ('new feature', 'refactor', 'redesign') and sharpening distinctiveness against adjacent planning/architecture skills.

DimensionReasoningScore

Specificity

Names three concrete deliverable areas — 'explicit requirements, technical design, and task planning before implementation' — which matches anchor 4 (several specific actions with minor gaps in coverage). It is not anchor 3 (only 1-2 concrete actions) nor anchor 5, since coverage is not comprehensive: outputs like spec documents, user stories, or staged execution coordination are not mentioned.

4 / 5

Completeness

Both halves are explicit: what ('explicit requirements, technical design, and task planning') and when ('Use when medium-to-large changes need... especially for multi-module work, unclear acceptance criteria, or architecture-heavy requests') with concrete trigger phrases. This matches anchor 5; anchor 4 does not fit because the 'when' clause is already explicit and specific rather than something that 'could be more explicit'.

5 / 5

Trigger Term Quality

Multiple fairly natural trigger phrases are present: 'medium-to-large changes', 'multi-module work', 'unclear acceptance criteria', 'architecture-heavy requests' — good coverage per anchor 4. It falls short of anchor 5 because common user variations such as 'new feature', 'refactor', 'redesign', or 'spec' are absent, but it exceeds anchor 3 since several distinct, relevant natural phrases are included.

4 / 5

Distinctiveness Conflict Risk

It carves a clear niche (spec-driven planning workflow) with mostly distinct triggers, fitting anchor 4. It is not anchor 5 because 'architecture-heavy requests' and 'medium-to-large changes' are somewhat broad and could overlap with general architecture or planning skills (including the referenced sibling skills), but it is more distinctive than anchor 3's 'could still overlap with similar skills'.

4 / 5

Total

17

/

20

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.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

TencentCloudBase/CloudBase-AI-Toolkit

Path

plugin/cloudbase/skills/spec-workflow/SKILL.md

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.