AgentSkillsCN

skill-task-model

技术规格（TASK）与用例的标准规范。

SKILL.md
--- frontmatter
name: skill-task-model
description: Standards for Technical Specifications (TASK) and Use Cases.
tier: 1
version: 1.1

TASK Model & Examples

Purpose: Defines the structure and quality standards for Technical Specifications to ensure unambiguous implementation.

1. Red Flags (Anti-Rationalization)

STOP if you are thinking:

  • •"I'll figure out the edge cases while coding" -> WRONG. define Alternative Scenarios NOW.
  • •"Acceptance Criteria: 'It works'" -> WRONG. Criteria must be binary (Pass/Fail) and specific.
  • •"I don't need actors, it's just a backend job" -> WRONG. "System" is an actor. "Cron Job" is an actor.

2. TASK Structure Rules

  • •Meta Information: Section 0 MUST include Task ID and Slug.
  • •Use Cases: Must be structured with Actors, Preconditions, Main Scenario, Alternative Scenarios, Postconditions, and Acceptance Criteria.
  • •Validation: Criteria must be verifiable.

3. Examples

✅ Example of Good Use Case

[!TIP] See examples/good_use_case.md for a complete, structured example.

❌ Example of Bad Use Case

[!TIP] See examples/bad_use_case.md for what NOT to do.

Rationalization Table

Agent ExcuseReality / Counter-Argument
"Writing scenarios takes too long"Rewriting code because of missed edge cases takes 3x longer.
"It's obvious what happens"Obvious to you now. Not obvious to the Code Reviewer or you in 2 weeks.

4. Resources

  • •examples/: Reference use cases.