open-fdd-lab

分类: 开发与编程 | 上传者: bbartling | 下载: 0 | 版本: v1.0（最新）

测试优先的 Open-FDD 实验室技能：外部工作台验证、前端/API 奇偶校验、BRICK+BACnet 验证、隔夜分类以及已确认产品缺陷的问题归档。

更新日志: Source: GitHub https://github.com/bbartling/open-fdd

目录结构

当前层级: tree/master/openclaw/

📁 assets/
- 📄 README.md 275 B
📁 bench/
- 📁 e2e/
  - 📄 1_e2e_frontend_selenium.py 78.6 KB
  - 📄 2_sparql_crud_and_frontend_test.py 72.1 KB
  - 📄 3_long_term_bacnet_scrape_test.py 40.8 KB
  - 📄 4_hot_reload_test.py 15.0 KB
  - 📄 5_ai_data_model_payload_test.py 14.1 KB
  - 📄 ai_modeling_pass.py 9.2 KB
  - 📄 automated_suite.py 4.3 KB
  - 📄 one_pass_runner.py 7.0 KB
  - 📄 README.md 2.7 KB
  - 📄 requirements-e2e.txt 279 B
- 📁 fake_bacnet_devices/
  - 📁 templates/
    
    📄 fake_bacnet.service.j2 497 B
  - 📄 ansible.cfg 547 B
  - 📄 fake_ahu_faults.py 11.3 KB
  - 📄 fake_vav_faults.py 7.2 KB
  - 📄 fault_schedule.py 3.1 KB
  - 📄 fix_dpkg.yml 1.1 KB
  - 📄 inventory.yml 934 B
  - 📄 playbook.yml 5.9 KB
  - 📄 README.md 7.1 KB
  - 📄 requirements.txt 17 B
- 📁 fixtures/
  - 📄 demo_site_llm_payload.json 20.1 KB
  - 📄 demo_site_llm_payload_malformed.json 185 B
  - 📄 demo_site_llm_payload_missing_site.json 567 B
- 📁 reports/
  - 📄 .gitkeep 0 B
- 📁 rules_lab/
  - 📄 README.md 590 B
- 📁 rules_reference/
  - 📄 ahu_fc1.yaml 929 B
  - 📄 ahu_fc10.yaml 771 B
  - 📄 ahu_fc11.yaml 684 B
  - 📄 ahu_fc12.yaml 892 B
  - 📄 ahu_fc13.yaml 791 B
  - 📄 ahu_fc14.yaml 1.2 KB
  - 📄 ahu_fc15.yaml 1.4 KB
  - 📄 ahu_fc16.yaml 778 B
  - 📄 ahu_fc2.yaml 889 B
  - 📄 ahu_fc3.yaml 891 B
  - 📄 ahu_fc4.yaml 758 B
  - 📄 ahu_fc5.yaml 865 B
  - 📄 ahu_fc6.yaml 914 B
  - 📄 ahu_fc7.yaml 766 B
  - 📄 ahu_fc8.yaml 839 B
  - 📄 ahu_fc9.yaml 942 B
  - 📄 ahu_rule_a.yaml 788 B
  - 📄 chiller_flow_fc2.yaml 528 B
  - 📄 chiller_pump_fc1.yaml 698 B
  - 📄 hp_discharge_cold_when_heating.yaml 758 B
  - 📄 README.md 843 B
  - 📄 sensor_bounds.yaml 1.9 KB
  - 📄 sensor_flatline.yaml 525 B
  - 📄 weather_gust_lt_wind.yaml 498 B
  - 📄 weather_rh_out_of_range.yaml 438 B
  - 📄 weather_temp_spike.yaml 488 B
  - 📄 weather_temp_stuck.yaml 546 B
- 📁 scripts/
  - 📄 ai_modeling_pass.py 4.6 KB
  - 📄 monitor_fake_fault_schedule.py 7.7 KB
- 📁 sparql/
  - 📄 01_platform_config.sparql 232 B
  - 📄 02_site_counts.sparql 523 B
  - 📄 02_sites_labels.sparql 360 B
  - 📄 03_equipment_labels_testbench.sparql 448 B
  - 📄 03_point_labels_testbench.sparql 472 B
  - 📄 04_bacnet_devices.sparql 231 B
  - 📄 05_brick_rule_mapping.sparql 642 B
  - 📄 06_polling_points_brick_type.sparql 723 B
  - 📄 07_count_triples.sparql 153 B
  - 📄 08_bacnet_telemetry_points.sparql 1.2 KB
  - 📄 09_graph_db_sync_counts.sparql 651 B
  - 📄 10_ahus.sparql 194 B
  - 📄 11_zones.sparql 175 B
  - 📄 12_building.sparql 452 B
  - 📄 13_vav_boxes.sparql 277 B
  - 📄 14_vavs_per_ahu.sparql 359 B
  - 📄 15_chillers.sparql 157 B
  - 📄 16_cooling_towers.sparql 169 B
  - 📄 17_boilers.sparql 155 B
  - 📄 18_central_plant.sparql 394 B
  - 📄 19_hvac_equipment.sparql 197 B
  - 📄 20_meters.sparql 316 B
  - 📄 21_points.sparql 168 B
  - 📄 22_class_summary.sparql 300 B
  - 📄 23_orphan_external_references.sparql 511 B
  - 📄 24_operator_site_context.sparql 1.3 KB
  - 📄 README.md 3.1 KB
📁 dashboard/
- 📄 favicon.ico 672 B
- 📄 index.html 20.5 KB
📁 docs/
- 📄 README.md 252 B
📁 logs/
- 📄 bootstrap-test-2026-03-25_09-00-30.txt 283.2 KB
- 📄 bootstrap-test-2026-03-25_12-19-03.txt 31.8 KB
📁 references/
- 📄 api_throttle.md 2.1 KB
- 📄 bench_operator_notes_2026-03-30.md 3.0 KB
- 📄 bootstrap_mcp_frontend.md 1.5 KB
- 📄 frontend_testing.md 668 B
- 📄 future_operator_clones.md 1.1 KB
- 📄 legacy_automated_testing.md 4.2 KB
- 📄 long_run_lab_pass.md 5.9 KB
- 📄 parity_issue_checklist.md 2.4 KB
- 📄 security_testing_scope.md 1022 B
- 📄 session_status_summary.md 4.7 KB
- 📄 skill_installation.md 1.3 KB
- 📄 testing_layers.md 3.4 KB
📁 reports/
- 📁 drafts/
  - 📄 issue-draft-openfdd-graph-drift-2026-03-24.md 2.5 KB
  - 📄 issue-draft-openfdd-parity-2026-03-24.md 2.2 KB
  - 📄 README.md 853 B
- 📄 overnight-bacnet-verification-template.md 947 B
- 📄 README.md 698 B
📁 scripts/
- 📄 capture_bootstrap_log.sh 525 B
- 📄 README.md 566 B
- 📄 verify_with_log.sh 377 B
📁 windows/
- 📄 run_daytime_short_suite.cmd 205 B
- 📄 run_midnight_suite.cmd 190 B
- 📄 run_overnight_bacnet.cmd 249 B
📄 HANDOFF_PROTOCOL.md 9.0 KB
📄 issues_log.md 15.3 KB
📄 README.md 4.5 KB
📄 SKILL.md 3.5 KB

SKILL.md

---
name: open-fdd-lab
description: Testing-first Open-FDD lab skill: external bench validation, frontend/API parity, BRICK+BACnet verification, overnight triage, and issue filing for confirmed product defects.
metadata: {"openclaw":{"homepage":"https://github.com/bbartling/open-fdd/tree/develop/openclaw"}}
user-invocable: true
---

# Open-FDD OpenClaw skill (testing-first)

This skill exists to test and evaluate a running Open-FDD system like an expert commissioning-minded web application tester and BRICK/BACnet data-model analyst.

Primary mission:
- verify frontend workflows
- verify backend/data-model behavior
- verify BACnet integration and add-to-model flows
- run overnight stability checks
- classify failures
- file high-quality bug reports for confirmed defects

Do not default to clone-first or repo-first workflows.

## Default posture

Assume Open-FDD is an externally running bench/deployment under test.

Start with runtime discovery:
1. active frontend URL
2. active backend URL
3. BACnet gateway target
4. auth context (`OFDD_API_KEY` validity)
5. model/SPARQL baseline behavior

Use UI behavior, API responses, SPARQL results, BACnet reads, and logs as primary evidence.
Use repo source/docs as supporting context unless local edits are explicitly requested.

## Current operational guardrails

- Current `FORBIDDEN: Invalid API key` behavior should be treated as auth/runtime-context drift until proven otherwise.
- Do not label auth drift as a confirmed product bug by default.
- Keep issue `#92` active for likely frontend/API parity defect tracking after auth is healthy.
- Do not delete or bury `#92` during cleanup work.

## Testing layers (required order of thinking)

1. Frontend/web app workflows and visible errors
2. Backend/API auth + data-model/SPARQL correctness
3. BACnet add-to-model and reference/read integrity
4. Overnight scrape/FDD/hot-reload stability
5. Future live-HVAC operator checks (when explicitly requested)

## Failure classification (required)

Every meaningful failure must be classified as one of:
- auth/launcher/env drift
- bench limitation
- frontend/API parity bug
- graph hygiene/model drift bug
- BACnet integration bug
- likely real Open-FDD product defect

File GitHub issues for confirmed product defects by default. Harness/runtime failures stay in `openclaw/issues_log.md` unless Ben explicitly requests issue tracking for harness debt too.

## Security phases

Track security as phased product hardening:
- auth robustness
- Caddy/proxy boundaries
- secrets handling
- attack-surface reduction
- roadmap-linked hardening tasks

Do not blend security hardening notes into unrelated defect triage without clear labeling.

## Workflow constraints

- Do not create/manage a second Open-FDD clone as baseline workflow.
- Do not treat coding changes as the primary goal.
- Keep issue/evidence trail in `openclaw/issues_log.md`.
- Use `openclaw/HANDOFF_PROTOCOL.md` for Cursor/OpenClaw handoff discipline.
- In the current operating model: **Cursor = product engineer, OpenClaw = tester**.
- When Cursor provides a commit SHA + issue IDs + acceptance criteria, retest the live bench against that SHA, do **no product-code edits**, and post evidence back to GitHub.
- Healthy auth preflight is required before drawing frontend/API parity conclusions.

## References to read first

1. `openclaw/HANDOFF_PROTOCOL.md`
2. latest dated section in `openclaw/issues_log.md`
3. `openclaw/references/testing_layers.md`
4. `openclaw/references/long_run_lab_pass.md`
5. `openclaw/references/api_throttle.md`

登录后下载/点赞/收藏 ❤ 115 | ★ 0

open-fdd-lab

目录结构

SKILL.md

举报内容

提示