菜单
Change impact scan for Python projects. When modifying function signatures, interfaces, data models, config structs, or shared modules, automatically grep all call sites and dependent files to prevent incomplete changes. Outputs a structured markdown impact report.
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。
Multi-model adversarial code review for agent outputs. Use when reviewing core logic changes, 5+ files, 200+ lines, or when user says "对抗审查/adversarial review/cross-model review". The reviewer agent inspects coder outputs through a cross-validation lens to catch single-model blind spots.
Pitfall recording and evolution tracking. When AI makes a mistake or user corrects output, record a PIT entry. Same-type PITs reaching 2+ trigger rule extraction; 4+ trigger skill crystallization. Use when user says "踩坑/pitfall/PIT/记录错误/犯错日志/evolve mistake/踩坑记录".
Use when developing, modifying, refactoring, or reviewing the task-orchestrator multi-agent orchestration engine. Covers design principles, architecture decisions, and development guardrails derived from the Multi-Agent Harness design reference. Triggered by changes to src/openagents_orchestration/, adding new orchestration features, or questions about orchestrator design.
Run a static code review over a target directory. Scans Python/JS/TS files for common issues (long functions, TODOs, missing docstrings, unused imports, security smells) and produces a structured markdown report.
Clean and transform tabular data (CSV/JSON). Performs deduplication, null-row removal, string trimming, and column normalization. Useful for ETL tasks that don't need pandas — uses only Python standard library.
Generate API documentation (markdown) from a Python module by AST parsing. Extracts module docstrings, function signatures, class hierarchies, and type hints. Pure Python, no LLM dependency.
| name | impact-scan-pipeline |
| description | Change impact scan for Python projects. When modifying function signatures, interfaces, data models, config structs, or shared modules, automatically grep all call sites and dependent files to prevent incomplete changes. Outputs a structured markdown impact report. |
Core: "Incomplete change" is a high-frequency AI coding bug — local perspective forgets global impact. This pipeline enforces global scanning.
| Type | Indicator |
|---|---|
| Signature change | Parameters, return values, method changes |
| Data model | Pydantic model field add/remove/modify |
| Config change | Config dataclass field add/remove/modify |
| Route change | API path, HTTP method, middleware chain changes |
This skill automatically executes the scan via run_openagent_skill.
Input parameters:
target_path (str, required) — The file that was changed.project_root (str, optional) — Project root for scanning. Default ..symbols (list[str], optional) — Specific symbol names to scan. If omitted, extracts functions/classes from the target file automatically.output_path (str, required) — Where to write the impact report.Scan strategy:
.py filesimport | call | attribute_access | string_refThe report has two sections per symbol:
| Section | Meaning | Action |
|---|---|---|
| Direct references | Import or call sites | Must sync modify |
| Indirect references | String matches, attribute access | Needs confirmation |
Show the impact report to the user, confirm scope before starting modifications.
# Change Impact Report
- **Target:** `src/openagents_orchestration/state_board.py`
- **Symbols scanned:** Budget, AgentState
## Symbol: `Budget`
### Direct references (must sync modify)
| File | Line | Type | Context |
|------|------|------|---------|
| `src/runner.py` | 89 | import | from state_board import Budget |
| `tests/test_state_board.py` | 15 | call | b = Budget(max_steps=10) |
### Indirect references (2 potential impacts)
| File | Line | Type | Context |
|------|------|------|---------|
| `docs/metrics.md` | 42 | string_ref | "Budget tracks token usage" |
| Error | Consequence | Prevention |
|---|---|---|
| Only change function, not call sites | ImportError or runtime bug | Step 2 grep scan |
| Change code but not config | Config drift | Step 3 config sync check |
| Start editing before scan completes | Miss associated changes | Step 4 confirm first |