feedback-loop当用户对输出给出学习或行为反馈(称赞/批评/达成共识/改主意)时,按反馈学习循环处理写入 feedback-log,重要反馈走完整落盘链(feedback-log→conclusions→MEMORY→验证可检索),保持立场一致不横跳。触发词:做得好/记住了/有问题/怎么又这样/记住/以后都这样(仅在指向本输出/行为的评价时生效)。
Install via ClawdBot CLI:
clawdbot install harrylabsj/feedback-loopGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Calls external URL not in known-safe list
https://github.com/openclaw/skills-feedback-loopAudited Apr 17, 2026 · audit v1.0
Generated Mar 20, 2026
A company uses the skill to collect explicit feedback (ratings and comments) after each support interaction and implicit feedback from session patterns like retries or early abandonment. This data is analyzed weekly to identify recurring issues, generate prioritized improvement suggestions for agent responses, and track the impact of implemented changes on customer satisfaction scores.
An educational platform integrates the skill to gather feedback on tutoring sessions, using explicit ratings for lesson helpfulness and implicit signals like completion rates or time spent. Analysis reveals trends in student engagement, leading to suggestions for refining content delivery and tracking improvements in learning outcomes over time.
An e-commerce business employs the skill to monitor user interactions with a shopping assistant, collecting feedback on product recommendations and checkout assistance. Implicit signals from session metrics help detect friction points, while analysis generates actionable suggestions to boost conversion rates and streamline the user journey.
A healthcare provider uses the skill to collect feedback on a virtual assistant handling appointment scheduling and basic inquiries. Explicit feedback assesses accuracy and empathy, while implicit data tracks completion rates. Analysis identifies areas for improvement in response clarity, with suggestions tracked to ensure compliance and enhance patient experience.
An organization deploys the skill to gather feedback on an HR bot answering policy and benefits questions. Employees provide explicit ratings, and implicit signals like correction attempts highlight confusing responses. Weekly analysis generates suggestions to update knowledge bases and improve communication, with tracking to measure resolution times and employee satisfaction.
Offer the skill as a cloud-based service with tiered pricing based on feedback volume and analysis features. Revenue comes from monthly or annual subscriptions, targeting businesses seeking continuous AI agent improvement without heavy infrastructure investment.
Provide custom integration and consulting services to help clients deploy and optimize the skill for their specific AI agents. Revenue is generated through project-based fees and ongoing support contracts, focusing on enterprises with complex feedback needs.
Offer a free version with basic feedback collection and CLI usage, while charging for advanced analytics, automated reporting, and priority support. Revenue streams from upgrades to premium plans, appealing to small teams and scaling businesses.
💬 Integration Tip
Start by integrating the CLI commands into existing agent workflows for manual feedback collection, then gradually automate implicit signal detection using the programming API to build a comprehensive feedback loop.
Scored Apr 19, 2026
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...
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
Global, namespaced learning memory for OpenClaw. Use when users correct output, set stable preferences, ask what was learned, ask for memory stats, or reques...
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...