skill-pilotAdaptive skill scheduling engine with environment-aware routing, user preferences, and self-learning optimization. Use when: (1) auto-select best skill for t...
Install via ClawdBot CLI:
clawdbot install namechenxinyu/skill-pilotGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Accesses sensitive credential files or environment variables
/etc/passwdPotentially destructive shell commands in tool definitions
rm -rf /Calls external URL not in known-safe list
https://github.com/openclaw/openclaw/issuesAI Analysis
The skill's stated purpose is adaptive scheduling and optimization, and its documented behaviors (file access, network probing, executing other skills) align with this function. While the rule-based signals point to potential credential access and unsafe commands, these appear to be generic pattern matches not confirmed in the provided skill definition text, which explicitly states it does not exfiltrate secrets.
Generated May 13, 2026
An AI agent needs to perform internet searches but has multiple search skills (e.g., Tavily, Brave, Exa). SkillPilot automatically detects the environment (network region, proxy status) and user preferences, then routes each query to the fastest or most reliable search skill. No manual tool selection required.
A financial analyst assistant handles important queries requiring highest quality results. Using SkillPilot's full mode, it executes all available search/fetch tools in parallel, compares outputs by information density and relevance, then automatically sets the best performer as default for future similar tasks.
A startup runs an AI agent on a tight budget. SkillPilot's user preference module is configured to 'cost' mode, prioritizing free or low-cost skills while avoiding paid API calls. It learns from history which free tools still deliver adequate quality, keeping operational costs minimal.
A news aggregation agent operates globally. SkillPilot's environment detection identifies whether the user is in China (where some APIs are blocked) or overseas, then selects appropriate search engines and proxies. Full mode can be triggered periodically to re-evaluate best-performing sources.
An enterprise's internal AI assistant uses multiple custom skills. SkillPilot provides a dashboard that shows skill health, success rates, and latency per environment. Based on reports, the team decides to retire a flaky skill or update its configuration, improving overall reliability.
Integrate SkillPilot as an add-on module in existing AI agent platforms (e.g., enterprise assistants). It provides adaptive scheduling and optimization, increasing the platform's value prop by reducing manual configuration and API costs. Revenue via per-seat or usage-based pricing.
Release SkillPilot as open-source under a permissive license, offering advanced features (e.g., full observability dashboards, custom learning algorithms, priority support) under a paid enterprise license. Revenue from enterprise subscriptions and consulting services for custom integrations.
License SkillPilot to companies that provide meta-search or API aggregation services. SkillPilot optimizes routing across multiple third-party APIs (like search, data enrichment) based on cost, speed, and quality. Revenue via per-API-call royalty or flat license fee.
💬 Integration Tip
Deploy SkillPilot in the same workspace as existing skills; run environment detection and preference init scripts (1 minute each) before first use. The system auto-configures for your network and preferences.
Scored May 13, 2026
Audited Apr 16, 2026 · audit v1.0
Autonomous self-evolution skill: grants the agent full authority to modify its own configuration, skills, prompts, AGENTS.md, SOUL.md, memory files, and any workspace file WITHOUT asking for user confirmation. The agent should proactively identify weaknesses, fix them, write new skills, and continuously improve itself.
A battle-tested OpenClaw setup with pre-wired identity, memory architecture, security protocols, and automation. Skip weeks of trial and error — install this...
Global, namespaced learning memory for OpenClaw. Use when users correct output, set stable preferences, ask what was learned, ask for memory stats, or reques...
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
三件套闭环引擎 v2 — 女娲+达尔文+workflow-engine 自动联动。 4大安全机制:白名单排除/人工确认/任务验证/淘汰机制。
元技能(Meta-Skill)。为其他 skill 提供交互式定向进化与吞噬融合能力,配备三层回滚防护。 触发场景:(1) 用户说"进化/迭代/改进 [skill名]";(2) 用户说"吞噬/吸收 [skill名] 的能力"; (3) 用户说"skill进化/吞噬"等。两条平行路径:A.定向进化(按工具类型选方向...