菜单
Token-efficient tracking for AI orchestration. CLI-first for status updates (~50 tokens), agent fallback for complex ops (~1KB). Use when: updating task status, querying blockers, creating progress files, validating phases.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Add non-text files to a person's artifacts folder. Use when saving images, documents, or other files related to someone. Trigger words: artifact, save image, add photo, attach file, store document.
AshAi extension guidelines for integrating AI capabilities with Ash Framework. Use when implementing vectorization/embeddings, exposing Ash actions as LLM tools, creating prompt-backed actions, or setting up MCP servers. Covers semantic search, LangChain integration, and structured outputs.
ASI Agent-O-Rama Skill
This skill should be used when solving hard questions, complex architectural problems, or debugging issues that benefit from GPT-5 Pro or GPT-5.1 thinking models with large file context. Use when standard Claude analysis needs deeper reasoning or extended context windows.
Personal assistant for daily routines, task management, and productivity
Coordinate Claude Code skills with OpenAI Codex autonomous workflows for end-to-end ATFT-GAT-FAN maintenance.
| name | artifact-tracking |
| description | Token-efficient tracking for AI orchestration. CLI-first for status updates (~50 tokens), agent fallback for complex ops (~1KB). Use when: updating task status, querying blockers, creating progress files, validating phases. |
Token-efficient tracking artifacts for AI agent orchestration.
| Operation | Command | Tokens |
|---|---|---|
| Mark complete | python scripts/update-status.py -f FILE -t TASK-X -s completed | ~50 |
| Batch update | python scripts/update-batch.py -f FILE --updates "T1:completed,T2:completed" | ~100 |
| Query pending | python scripts/query_artifacts.py --status pending | ~50 |
| Validate | python scripts/validate_artifact.py -f FILE | ~50 |
Scripts location: .claude/skills/artifact-tracking/scripts/
For complex operations requiring judgment:
| Operation | Agent | When to Use |
|---|---|---|
| CREATE file | artifact-tracker | New phase, need template |
| UPDATE complex | artifact-tracker | Blockers with context, decisions |
| QUERY synthesis | artifact-query | Cross-phase analysis, handoffs |
| VALIDATE quality | artifact-validator | Pre-completion checks |
Agent invocation:
Task("artifact-tracker", "Create Phase 2 progress for auth-overhaul PRD")
Task("artifact-query", "Show all blocked tasks in auth-overhaul phases 1-3")
| Type | Location | Limit |
|---|---|---|
| Progress | .claude/progress/[prd]/phase-N-progress.md | ONE per phase |
| Context | .claude/worknotes/[prd]/context.md | ONE per PRD |
| Bug fixes | .claude/worknotes/fixes/bug-fixes-YYYY-MM.md | ONE per month |
| Observations | .claude/worknotes/observations/observation-log-MM-YY.md | ONE per month |
Policy: .claude/specs/doc-policy-spec.md
---
type: progress
prd: "prd-name"
phase: 2
status: in_progress
progress: 40
tasks:
- id: "TASK-2.1"
status: "pending" # pending|in_progress|completed|blocked
assigned_to: ["agent-name"] # REQUIRED for orchestration
dependencies: [] # REQUIRED for orchestration
model: "opus" # Optional: opus|sonnet|haiku
parallelization:
batch_1: ["TASK-2.1", "TASK-2.2"] # Run parallel
batch_2: ["TASK-2.3"] # After batch_1
---
| Operation | Traditional | Optimized | Savings |
|---|---|---|---|
| Task list | 25KB | 2KB | 92% |
| Query blockers | 75KB | 3KB | 96% |
| Status update | 25KB | 50 bytes | 99.8% |
./creating-artifacts.md./updating-artifacts.md./querying-artifacts.md./validating-artifacts.md./orchestration-reference.md./best-practices.md./common-patterns.md./format-specification.md./templates/./schemas/