Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Upgrade OpenClaw and discover new features, hooks, and config improvements. Use when: user explicitly says "upgrade openclaw", "update openclaw", "check for...
Upgrade OpenClaw and discover new features, hooks, and config improvements. Use when: user explicitly says "upgrade openclaw", "update openclaw", "check for...
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
Update, diff, audit, propose. Every new feature surfaced. Nothing missed.
On first run, check settings.json in this skill's directory. If subagentModel not set, ask: "Which model for upgrade sub-agents? (e.g., claude-sonnet-4-6, deepseek-chat). Note: external providers will receive config data." Save to settings.json: { "subagentModel": "anthropic/claude-sonnet-4-6" }
Before touching anything: PRE_VERSION=$(openclaw --version | grep -oP '\d{4}\.\d+\.\d+') echo "$PRE_VERSION" Save PRE_VERSION β needed for changelog diffing in Step 3.
openclaw update If dirty working tree, stash first: cd "$(openclaw --version 2>&1 | grep -oP '(?<=\().*?(?=\))' || echo ~/openclaw)" git stash --include-untracked -m "pre-update stash" && openclaw update Record new version: POST_VERSION=$(openclaw --version | grep -oP '\d{4}\.\d+\.\d+') If PRE_VERSION == POST_VERSION, report "Already up to date" and skip to Step 5 (audit only).
The changelog lives locally at ~/openclaw/CHANGELOG.md after update. Versions delimited by ## YYYY.x.x headers. Extract only entries between old and new version: awk "/^## $POST_VERSION/,/^## $PRE_VERSION/" ~/openclaw/CHANGELOG.md Then filter by relevance to this setup: Read current config via gateway config.get to identify enabled channels/plugins From the changelog delta, keep entries matching: Enabled channels (e.g., Telegram β skip LINE/Discord/Feishu/etc. unless enabled) Core agent/gateway/cron/tools/memory/security changes (always relevant) ACP/sessions/subagent changes (if ACP enabled) Breaking changes (always relevant) Discard entries for disabled channels, iOS/macOS app changes, and platforms not in use Categorize kept entries into: Features | Fixes | Security | Breaking Changes
Fetch the live schema and current config: Schema: gateway config.schema Current: gateway config.get Compare systematically: Walk schema properties tree recursively For each schema key path, check if it exists in current config New/unset options = schema keys not present in current config (excluding keys with sensible defaults) Focus on actionable fields β ones where setting a non-default value provides clear benefit Flag fields from the changelog delta (new in this version) separately as "New in this release" Present as a table: | Config Path | Type | Default | Description | New? |
openclaw hooks list --json openclaw doctor --non-interactive clawhub update --all --dry-run 2>&1 Collect: Hooks: any hooks available but not enabled Doctor: all warnings and recommendations Skills: any ClawHub updates available
Never apply without explicit user approval. On approval, apply changes via: Config changes: gateway config.patch Hook enablement: openclaw hooks enable <hook> Skill installs: clawhub install <skill> After applying, pop any git stash: cd ~/openclaw && git stash list | grep -q "pre-update stash" && git stash pop
Write upgrade state to state.json in this skill's directory: { "lastUpgrade": { "from": "2026.3.2", "to": "2026.3.3", "timestamp": "2026-03-05T06:50:00Z", "featuresProposed": ["telegram-streaming", "pdf-tool", ...], "featuresApplied": ["telegram-streaming", ...], "doctorApplied": ["entrypoint-fix", ...] } } This prevents re-proposing on repeated runs and enables "what changed since last upgrade" queries.
Agent frameworks, memory systems, reasoning layers, and model-native orchestration.
Largest current source with strong distribution and engagement signals.