AgentSkillsCN

new-ability-high-flow

高层次能力脚手架的分步流程。关键词:能力、高层次、流程。

SKILL.md
--- frontmatter
name: new-ability-high-flow
description: "Step-by-step flow for high-level ability scaffolding. Keywords: ability, high-level, flow."

Flow: New High-Level Ability

This doc contains the step-by-step flow. For inputs, tools, outputs, and safety, see /.system/skills/ssot/repo/scaffolding/new-ability-high/SKILL.md.


Step-by-Step Flow (AI + Human)

Step 1: Collect Parameters (AI -> Human)

AI asks human for:

  1. Ability ID (validate: unique, follows naming convention)
  2. Summary description
  3. Ability type (workflow/agent)
  4. Entrypoints (which ABILITY.md files should list this)
  5. Type-specific implementation details

Human checkpoint: Confirm parameters before proceeding.

Step 2: Validate Inputs (AI)

AI validates:

  • id uniqueness in existing registries
  • All entrypoints exist and have ABILITY.md with marker blocks
  • Implementation reference validity
  • Hook bindings reference valid hooks (if specified)

Step 3: Preview Changes (AI)

AI runs orchestrator with --dry-run:

bash
python scripts/devops/scaffold/new_ability_high.py   --id "<ability_id>"   --summary "<summary>"   --type <type>   --entrypoints "/modules/integration"   --dry-run

AI presents planned_changes to human.

Human checkpoint: Review and approve planned changes.

Step 4: Execute Registration (AI)

AI runs orchestrator without --dry-run:

bash
python scripts/devops/scaffold/new_ability_high.py   --id "<ability_id>"   --summary "<summary>"   --type <type>   --entrypoints "/modules/integration"

Step 5: Regenerate ABILITY.md Indexes (AI)

The orchestrator automatically calls:

bash
python scripts/devops/registry/abilities.py --scope integration --regenerate

Step 6: Verify Results (AI)

AI verifies:

  • Registry entry exists in /.system/registry/high-level/
  • ABILITY.md indexes are updated
  • abilities.py --check passes

Step 7: Record in Workdocs (AI)

AI records scaffold execution:

  • Task ID, timestamp, parameters
  • Registry entry created
  • ABILITY.md files updated
  • Follow-up: implement workflow steps or agent logic if not yet done