Files
DBAdmin/archive/NASA/openclaw_workspace.md
T
2026-05-18 06:40:19 +00:00

275 lines
8.4 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
## 1) Ustal ROOT\_PATH (gdzie ma powstać system)
Najbezpieczniej jest trzymać to w **workspace OpenClaw**, bo sandbox i polityki narzędzi (fs) zwykle są tam gotowe. [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli/sandbox), [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli)
### Opcja rekomendowana (workspace OpenClaw)
```bash
export ROOT_PATH="$HOME/.openclaw/workspace/missions"
```
### Opcja alternatywna (Twoje `.org` / JD)
Jeśli chcesz podpiąć pod Johnny Decimal, użyj np.:
```bash
export ROOT_PATH="$HOME/.org/missions"
```
> Tylko pamiętaj: sandbox musi mieć prawo zapisu do tej ścieżki — sprawdzisz to w kroku 3. [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli/sandbox)
***
## 2) Uruchom wszystko w osobnym profilu (polecane)
Profil `missionctl` odseparuje konfigurację od reszty. OpenClaw wspiera profile flagą `--profile`. [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli)
```bash
openclaw --profile missionctl gateway start
openclaw --profile missionctl gateway status
```
***
## 3) Sprawdź sandbox i dostęp do FS (krytyczne)
To pokaże, czy agent może pisać do `ROOT_PATH` i jakie ma ograniczenia. [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli/sandbox)
```bash
openclaw --profile missionctl sandbox explain
```
Jeśli wyjdzie, że sandbox nie ma dostępu do Twojej ścieżki poza workspace — najprościej przenieść `ROOT_PATH` do `~/.openclaw/workspace/…`.
***
## 4) (Opcjonalnie) Ustaw profil narzędzi tak, by mieć operacje plikowe
OpenClaw ma polityki narzędzi i profile; `openclaw config set` to standardowy sposób ustawiania configu. [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli), [\[deepwiki.com\]](https://deepwiki.com/openclaw/openclaw/6.2-tool-security-and-sandboxing)
```bash
openclaw --profile missionctl config set tools.profile '"coding"'
```
***
## 5) BOOTSTRAP: utwórz cały system z diagramu (foldery + pliki + config)
Poniższe polecenie uruchamia **jednego “turna” agenta** przez OpenClaw, który:
* tworzy strukturę katalogów `_MCC`, `APOLLO`, `HUBBLE`, `ARTEMIS`
* zakłada sekcje i katalogi `missions/`
* zapisuje pliki startowe (Operating Manual, Routing Rules, indeksy)
* zapisuje templatey
* zapisuje `_MCC/openclaw.missions.json` (mapowanie callsign → folder + agenci)
`openclaw agent` jest wprost w CLI. [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli)
```bash
openclaw --profile missionctl agent --message "
BOOTSTRAP Mission Management System.
ROOT_PATH=$ROOT_PATH
1) Create directories:
- $ROOT_PATH/_MCC/50_templates
- $ROOT_PATH/APOLLO/CORE_TECH/missions
- $ROOT_PATH/APOLLO/FLIGHT_CODE/missions
- $ROOT_PATH/APOLLO/VERIFICATION/missions
- $ROOT_PATH/APOLLO/99_archive
- $ROOT_PATH/HUBBLE/CONTENT/missions
- $ROOT_PATH/HUBBLE/CREATIVE/missions
- $ROOT_PATH/HUBBLE/99_archive
- $ROOT_PATH/ARTEMIS/EXPERIMENTS/missions
- $ROOT_PATH/ARTEMIS/TELEMETRY/missions
- $ROOT_PATH/ARTEMIS/GROUND_CREW/missions
- $ROOT_PATH/ARTEMIS/99_archive
2) Write MCC files (overwrite if exist):
A) $ROOT_PATH/_MCC/00_operating-manual.md
# Operating Manual — Mission System
## Governance
- PROGRAM LEAD: vision, strategy, final decisions
- MISSION CONTROL (MCC): triage, routing, execution, WIP, closure
## Missions
- APOLLO: Flight Systems (Core Tech, Flight Code, Verification)
- HUBBLE: Mission Story (Mission Content, Creative Studio)
- ARTEMIS: Mission Outcomes (Experiments, Telemetry, Ground Crew)
## Packet standard
Each packet contains:
00_brief.md, 10_worklog.md, 20_artifacts/, 30_results.md, 40_decision.md, 90_links.md
## Modes
SIM (default) vs FLIGHT (requires SIM→FLIGHT checklist)
## Closure
PROCEED | ITERATE | HOLD | SCRUB
B) $ROOT_PATH/_MCC/40_routing-rules.md
# Routing Rules (MCC)
## Call-signs → folders
APOLLO-CORE → APOLLO/CORE_TECH
APOLLO-CODE → APOLLO/FLIGHT_CODE
APOLLO-VERIFY → APOLLO/VERIFICATION
HUBBLE-CONTENT → HUBBLE/CONTENT
HUBBLE-CREATIVE → HUBBLE/CREATIVE
ARTEMIS-EXP → ARTEMIS/EXPERIMENTS
ARTEMIS-TLM → ARTEMIS/TELEMETRY
ARTEMIS-GROUND → ARTEMIS/GROUND_CREW
## Default mode
SIM is default. FLIGHT requires checklist + approvals.
C) Empty indices:
- $ROOT_PATH/_MCC/10_backlog.md with '# Backlog (MCC)'
- $ROOT_PATH/_MCC/20_active-missions.md with '# Active Missions (MCC)'
- $ROOT_PATH/_MCC/30_decisions.md with '# Decisions (MCC)'
- $ROOT_PATH/_MCC/90_archive-index.md with '# Archive Index (MCC)'
3) Write templates:
- $ROOT_PATH/_MCC/50_templates/task-brief.md
- $ROOT_PATH/_MCC/50_templates/sim-flight-checklist.md
- $ROOT_PATH/_MCC/50_templates/post-mission-report.md
- $ROOT_PATH/_MCC/50_templates/decision-record.md
Contents:
(task-brief.md)
# [CALLSIGN] Task Brief — <title>
Routing: [CALLSIGN]
Agent: <AgentName>
Mode: SIM | FLIGHT
Priority: P0 | P1 | P2
## Context
## Goal
## Deliverables
## Constraints
## Acceptance Criteria
## Risks
(sim-flight-checklist.md)
# SIM → FLIGHT Checklist
- Acceptance criteria met
- Artifacts linked in 90_links.md
- Risks reviewed
- Verification done (if needed)
- Telemetry plan ready (if needed)
- MCC GO
- PROGRAM LEAD GO (P0/P1)
(post-mission-report.md)
# Post-Mission Report — <title>
Packet:
Decision:
Summary:
Artifacts:
Measurements:
Learnings:
Next Steps:
(decision-record.md)
# Decision Record — <title>
Decision: PROCEED | ITERATE | HOLD | SCRUB
Date:
Owner:
Rationale:
Trade-offs:
Follow-up:
4) Write mapping config:
$ROOT_PATH/_MCC/openclaw.missions.json
Use this exact JSON, but with root_path = ROOT_PATH:
{
\"root_path\": \"$ROOT_PATH\",
\"wip_limit\": 5,
\"modes\": { \"default\": \"SIM\", \"flight_requires_checklist\": true, \"flight_requires_program_lead_for_priority\": [\"P0\",\"P1\"] },
\"missions\": {
\"APOLLO\": {
\"label\": \"FLIGHT SYSTEMS\",
\"sections\": {
\"APOLLO-CORE\": { \"folder\": \"APOLLO/CORE_TECH\", \"agents\": [\"Anvil\",\"Cipher\"] },
\"APOLLO-CODE\": { \"folder\": \"APOLLO/FLIGHT_CODE\", \"agents\": [\"Pixel\",\"Sentry\"] },
\"APOLLO-VERIFY\": { \"folder\": \"APOLLO/VERIFICATION\", \"agents\": [\"Inspector\"] }
}
},
\"HUBBLE\": {
\"label\": \"MISSION STORY\",
\"sections\": {
\"HUBBLE-CONTENT\": { \"folder\": \"HUBBLE/CONTENT\", \"agents\": [\"Rex\",\"Sage\",\"Echo\",\"Clip\"] },
\"HUBBLE-CREATIVE\": { \"folder\": \"HUBBLE/CREATIVE\", \"agents\": [\"Nebula\",\"Nova\"] }
}
},
\"ARTEMIS\": {
\"label\": \"MISSION OUTCOMES\",
\"sections\": {
\"ARTEMIS-EXP\": { \"folder\": \"ARTEMIS/EXPERIMENTS\", \"agents\": [\"Scout\",\"Herald\"] },
\"ARTEMIS-TLM\": { \"folder\": \"ARTEMIS/TELEMETRY\", \"agents\": [\"Forge\",\"Pulse\"] },
\"ARTEMIS-GROUND\": { \"folder\": \"ARTEMIS/GROUND_CREW\", \"agents\": [\"Beacon\",\"Link\",\"Vibe\"] }
}
}
},
\"packet\": {
\"name_format\": \"YYYY-MM-DD__CALLSIGN__slug\",
\"required_files\": [\"00_brief.md\",\"10_worklog.md\",\"30_results.md\",\"40_decision.md\",\"90_links.md\"],
\"artifact_dir\": \"20_artifacts\"
}
}
5) After completion, print a short summary of what you created and where.
"
```
***
## 6) Diagnostyka, jeśli coś “nie zadziała”
### Podgląd logów gateway (na żywo)
```bash
openclaw --profile missionctl logs --follow
```
`openclaw logs --follow` to oficjalny sposób podglądu logów Gateway. [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli)
### Szybki status całego systemu
```bash
openclaw --profile missionctl status --deep
```
Polecenie `openclaw status` (w tym `--deep`) jest w CLI i służy do diagnostyki. [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli)
***
## 7) (Bonus) Stwórz pierwszy Mission Packet testowy
Po bootstrapie możesz od razu sprawdzić, że wszystko działa — np. utworzyć pakiet w `ARTEMIS-TLM`:
```bash
openclaw --profile missionctl agent --message "
Utwórz Mission Packet w $ROOT_PATH dla:
CALLSIGN: ARTEMIS-TLM
Owner: Pulse
Mode: SIM
Slug: system-check
Folder: YYYY-MM-DD__ARTEMIS-TLM__system-check w ARTEMIS/TELEMETRY/missions/
W środku: 00_brief.md, 10_worklog.md, 20_artifacts/, 30_results.md, 40_decision.md, 90_links.md
Dopisz wpis do $ROOT_PATH/_MCC/20_active-missions.md ze statusem IN_PROGRESS.
"
```
***
# Dwa szybkie pytania (żeby dopasować to perfekcyjnie pod Twój Linux)
1. Jaki jest u Ciebie **ROOT\_PATH docelowo**: w `~/.openclaw/workspace/…` czy w Twoim `.org` (Johnny Decimal)?
2. Czy OpenClaw działa w Docker sandboxie czy “host mode”? (Wynik `openclaw sandbox explain` powie — jeśli wkleisz 23 linie, dopasuję zalecenia 1:1.) [\[docs.openclaw.ai\]](https://docs.openclaw.ai/cli/sandbox)