{"@context":"https://schema.org","@graph":[{"@type":"SoftwareApplication","@id":"https://hol.org/registry/plugins/domitorai%2Fscarwise#plugin","name":"Scarwise","description":"AI agents forget everything when a session ends: they call a tool the wrong way, find what works – and next session repeat the same mistake.\n\n**Scarwise** sits between your agents and their tools, on your computer. When a call fails and the agent then succeeds another way, it keeps a lesson; when any agent is about to repeat it, the call is stopped before it runs and the agent is told what worked. A tool that can never work here (missing key, program or app) is explained after the first failure. No rules to write.\n\n**Measured** on 152 real MCP servers (Qwen Code, OpenCode; same task in 3 new sessions): same mistake repeated **−48%**, failed calls **−33%**, error tokens **−37%**. MCPMark benchmark: failed calls **−27%**.\n\n**Works with** Claude Code, Codex, Gemini CLI, Cursor, Copilot, Kiro, Qwen Code, OpenCode, Claude.ai/ChatGPT (MCP proxy) and your own apps.\n\n**Install:** `npm install -g scarwise` (connects your agents by itself).\n\nLessons stay local; cleaned lessons are shared by default (`scarwise share off`). Free, closed source.","url":"https://hol.org/registry/plugins/domitorai%2Fscarwise","applicationCategory":"Tools & Integrations","softwareVersion":"v0.1.8","creator":"DomitorAI","codeRepository":"https://github.com/DomitorAI/scarwise","softwareRequirements":["Works with Standalone"],"sameAs":["https://github.com/DomitorAI/scarwise","https://github.com/DomitorAI/scarwise"],"image":["https://hol.org/api/registry/plugin-media/a2b4866f-b80b-4b0c-843e-a81c633d5233","https://img.shields.io/endpoint?url=https%3A%2F%2Fhol.org%2Fapi%2Fregistry%2Fbadges%2Fplugin%3Fslug%3Ddomitorai%252Fscarwise%26metric%3Dtrust%26style%3Dflat","https://img.shields.io/endpoint?url=https%3A%2F%2Fhol.org%2Fapi%2Fregistry%2Fbadges%2Fplugin%3Fslug%3Ddomitorai%252Fscarwise%26metric%3Dsecurity%26style%3Dflat"],"additionalProperty":[{"@type":"PropertyValue","name":"HOL Trust Score","value":79},{"@type":"PropertyValue","name":"Security Score","value":80},{"@type":"PropertyValue","name":"Security Result","value":"Needs review"},{"@type":"PropertyValue","name":"Skills Bundled","value":0},{"@type":"PropertyValue","name":"MCP Servers Bundled","value":0},{"@type":"PropertyValue","name":"Apps Bundled","value":0},{"@type":"PropertyValue","name":"Latest Launch","value":"Scarwise"},{"@type":"PropertyValue","name":"Launch Tagline","value":"Stops AI agents from repeating the same tool mistake in every new session"},{"@type":"PropertyValue","name":"Known limitation","value":"No license declared in the manifest."}]},{"@type":"ItemList","@id":"https://hol.org/registry/plugins/domitorai%2Fscarwise#related-alternatives","name":"Alternatives in Tools & Integrations","itemListOrder":"https://schema.org/ItemListOrderDescending","numberOfItems":4,"itemListElement":[{"@type":"ListItem","position":1,"name":"AI Command Center","url":"https://hol.org/registry/plugins/hredo%2Fai-command-center"},{"@type":"ListItem","position":2,"name":"DC Hub","url":"https://hol.org/registry/plugins/dc-hub%2Fdchub"},{"@type":"ListItem","position":3,"name":"MetaMynd AgentSafe Guard – Runtime Governance for AI Agents","url":"https://hol.org/registry/plugins/metamynd%2Fagentsafe-guard"},{"@type":"ListItem","position":4,"name":"MetaMynd MCP Gateway – AI Agent Security & Governance","url":"https://hol.org/registry/plugins/metamynd%2Fmetamynd-mcp-gateway"}]}]}