Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Tencent SkillHub Β· AI
Escalate
Auto-learns when to handle autonomously vs pause for human input. Grows trust over time, stays conservative until patterns confirm.
skill openclawclawhub Free
Auto-learns when to handle autonomously vs pause for human input. Grows trust over time, stays conservative until patterns confirm.
β¬ 0 downloads β
0 stars Unverified but indexed
Install for OpenClaw
Quick setup
- Download the package from Yavira.
- Extract the archive and review SKILL.md first.
- Import or place the package into your OpenClaw setup.
Package facts
- Download mode
- Yavira redirect
- Package format
- ZIP package
- Source platform
- Tencent SkillHub
- What's included
- SKILL.md, boundaries.md, memory-template.md, migration.md, patterns.md, setup.md
Validation
- Use the Yavira download entry.
- Review SKILL.md after the package is downloaded.
- Confirm the extracted package contains the expected setup assets.
Install with your agent
Agent handoff
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
New install
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.
Upgrade existing
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.
Trust & source
Release facts
- Source
- Tencent SkillHub
- Verification
- Indexed source record
- Version
- 1.0.2
Provenance
- Publisher
- ivangdavila
- Source page
- View original listing
- Canonical URL
- Open canonical page
Documentation
Architecture
Escalation state lives in ~/escalate/. If that folder is missing or empty, run setup.md. ~/escalate/ βββ memory.md # Stable activation rules, escalation posture, and saved boundaries βββ decisions.md # Recent escalation calls, corrections, and trust updates βββ domains/ # Optional domain-specific overrides (code, ops, comms)
When to Use
Use when the user wants the agent to decide what can be handled autonomously, what should be proposed first, and what always needs explicit approval. Turn this on for agents that draft, edit, research, or operate proactively and need a durable ask-vs-act policy across sessions.
Quick Reference
TopicFileSetup guidesetup.mdMemory templatememory-template.mdMigration guidemigration.mdHard boundariesboundaries.mdPattern guidepatterns.md
1. Move Fast on Safe Internal Work
Research, drafts, formatting, and reversible local edits should not turn into permission theater. If the action is low-risk and clearly aligned with precedent, act or act-then-inform. Save confirmed low-risk autonomy in memory so the same task does not need fresh approval every time.
2. Slow Down on External or Irreversible Impact
Money, deletion, deployment, public communication, approvals, and third-party consequences stay escalated unless the user set a very explicit exception. If the downside is asymmetric, ask before acting even when the path feels obvious. A fast, sharp escalation is better than a silent overstep.
3. Learn from Explicit Corrections, Not Vibes
"Just do that next time" promotes autonomy only for the matching context. "Ask me first on this" demotes autonomy immediately for the matching context. Silence is not consent and habit is not policy until the user makes it clear.
4. Match the Boundary to the Context
Same verb does not mean same risk: a small local refactor is not a production rewrite. Judge by reversibility, blast radius, external visibility, and cost. Prefer domain-specific overrides when the same user behaves differently in code, comms, and operations.
5. Escalate with a Recommendation
When approval is needed, bring the best option instead of a blank question. Keep the escalation short: what changed, why it matters, and the recommended move. The goal is confidence and speed, not bureaucratic caution.
6. Keep Workspace Routing Non-Destructive
Use setup to propose small additions to the workspace AGENTS file and SOUL file; never replace whole sections. Show the exact snippet before editing and preserve existing workspace language. Route escalation behavior through the workspace clearly enough that the agent knows exactly which escalate files to read before risky work.
7. Prefer Durable Defaults over Repeated Friction
Once the user confirms a safe pattern, reuse it aggressively in matching situations. Keep hard boundaries stable and explicit. If trust drops after a bad call, tighten the rule immediately and write it down.
Common Traps
TrapWhy It FailsBetter MoveAsking before every tiny internal actionFeels slow and timidAct on reversible local work once precedent existsTreating "just do it" as universal permissionOver-generalizes trustScope the grant to the matching action, stakes, and domainUsing the same threshold for code and external commsRisk profile changes by domainStore domain overrides in ~/escalate/domains/Escalating without a recommendationCreates decision fatigueOffer the best option and one-line rationaleEditing the workspace AGENTS or SOUL file wholesaleBreaks workspace identityAdd a small snippet and preserve everything else
Data Storage
Local state lives in ~/escalate/: stable escalation rules and activation preferences in ~/escalate/memory.md recent calls, corrections, and trust updates in ~/escalate/decisions.md optional domain-specific overrides in domains/ The packaged guides boundaries.md and patterns.md stay in the skill itself and act as references, not as the user's live memory.
Security & Privacy
This skill stores local escalation notes in ~/escalate/. It may read workspace steering files such as the AGENTS file and SOUL file to align the ask-vs-act policy. It may suggest small non-destructive edits to those files during setup, but it must show the snippet and wait for explicit approval before any write. It does not send messages, spend money, delete data, deploy, or approve legal terms without explicit approval. It never modifies its own SKILL.md.
Related Skills
Install with clawhub install <slug> if user confirms: self-improving - Learn reusable execution lessons from corrections and reflection proactivity - Push useful next steps without overstepping learned boundaries decide - Turn repeated choices into clearer decision rules memory - Keep durable user context and continuity across sessions
Feedback
If useful: clawhub star escalate Stay updated: clawhub sync
Category context
Agent frameworks, memory systems, reasoning layers, and model-native orchestration.
Source: Tencent SkillHub
Largest current source with strong distribution and engagement signals.
Package contents
Included in package
6 Docs
- SKILL.md
- boundaries.md
- memory-template.md
- migration.md
- patterns.md
- setup.md