{ "schemaVersion": "1.0", "item": { "slug": "rune", "name": "Rune - Self-Improving AI Memory", "source": "tencent", "type": "skill", "category": "AI ๆ™บ่ƒฝ", "sourceUrl": "https://clawhub.ai/TheBobLoblaw/rune", "canonicalUrl": "https://clawhub.ai/TheBobLoblaw/rune", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/rune", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=rune", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "CHANGELOG.md", "COMMON-MISTAKES.md", "INTEGRATION-GUIDE.md", "README.md", "SECURITY.md", "SKILL.md" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "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." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "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. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T16:55:25.780Z", "expiresAt": "2026-05-07T16:55:25.780Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/rune" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/rune", "agentPageUrl": "https://openagent3.xyz/skills/rune/agent", "manifestUrl": "https://openagent3.xyz/skills/rune/agent.json", "briefUrl": "https://openagent3.xyz/skills/rune/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "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." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "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. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "Rune - Persistent AI Memory System", "body": "Rune gives your OpenClaw agent persistent, intelligent memory that gets better over time. No more burning tokens on static context files or forgetting important information between sessions." }, { "title": "๐Ÿง  Smart Memory Management", "body": "Dynamic Context Injection: AI selects only relevant facts for each conversation\nAccess Pattern Learning: Frequently used facts get prioritized\nForgetting Curves: Unused facts naturally fade like human memory\nMemory Consolidation: Similar facts get merged, verbose ones compressed" }, { "title": "๐ŸŽฏ Session Intelligence", "body": "Interaction Style Detection: Learns if you prefer brainstorm vs deep-work vs debug modes\nBehavioral Pattern Analysis: Tracks your work patterns and preferences over time\nProactive Memory: Volunteers relevant context unprompted (\"last time you worked on this...\")" }, { "title": "๐Ÿ“‹ Project Autopilot", "body": "Smart Task Recommendations: \"What should I work on next?\" with priority scoring\nBlocker Detection: Identifies stuck projects that need intervention\nProject Health Scoring: 0.0-1.0 health scores based on activity and progress" }, { "title": "๐Ÿ“ข Intelligent Notifications", "body": "Priority Classification: Critical/High/Medium/Low/FYI with context analysis\nSmart Timing: Respects quiet hours, batches low-priority updates\nChannel Routing: DM for urgent, Discord for projects, digest for FYI" }, { "title": "๐Ÿ”„ Self-Improvement Loop", "body": "Pattern Detection: \"Forgot to use X skill 3 times\" โ†’ automatic escalation\nPerformance Tracking: Measurable improvement over time\nSkill Usage Analysis: Which skills you use vs neglect" }, { "title": "About Rune vs rune", "body": "Rune is the OpenClaw skill name. rune is the CLI tool name. Think of Rune as the \"skill package\" and rune as the \"command-line interface\" - like how the git skill package provides the git CLI.\n\nSkill name: rune (what you install via ClawHub)\nCLI command: rune (what you run in terminal)\nRepository: https://github.com/TheBobLoblaw/rune" }, { "title": "Installation Disclosure", "body": "โš ๏ธ What This Installation Does:\n\nThe Rune skill installation will:\n\nCreate directories: ~/.openclaw/ and subdirectories\nInstall globally: rune CLI via npm (requires npm dependencies)\nCreate database: SQLite database at ~/.openclaw/memory.db\nModify files: Appends integration lines to existing ~/.openclaw/workspace/HEARTBEAT.md\nAdd session hooks: Automatic memory integration for OpenClaw sessions\n\nBefore installing:\n\nBack up HEARTBEAT.md if it contains important data\nReview package.json dependencies if security is critical\nConsider using local models (Ollama) instead of cloud APIs for privacy" }, { "title": "Installation", "body": "# Via ClawHub (recommended)\nclawhub install rune\n\n# Manual installation\ngit clone https://github.com/TheBobLoblaw/rune\ncd rune\nnpm install --production\nnpm install -g ." }, { "title": "Quick Start", "body": "# Initialize memory system\nrune stats\n\n# Add your first fact\nrune add person cory.name \"Cory - my human user\"\n\n# Generate context for a conversation \nrune context \"Let's work on the website\"\n\n# Get task recommendations\nrune next-task\n\n# Weekly self-review\nrune self-review --days 7" }, { "title": "Memory Management", "body": "rune add - Store a fact\nrune search - Find facts\nrune recall - Smart multi-source recall\nrune inject - Generate context file for agent" }, { "title": "Intelligence Features", "body": "rune context - Dynamic context for message\nrune score - Relevance scoring\nrune proactive - Volunteer relevant context\nrune session-style - Detect interaction style" }, { "title": "Project Management", "body": "rune project-state - Track project phases/blockers\nrune next-task - Smart task picker\nrune stuck-projects - Find blocked work" }, { "title": "Advanced Features", "body": "rune temporal \"last Tuesday\" - Time-based queries\nrune consolidate - Memory optimization\nrune forget - Apply forgetting curves\nrune pattern-analysis - Detect behavioral patterns" }, { "title": "Heartbeat Integration", "body": "Add to your HEARTBEAT.md:\n\n# Memory maintenance\nrune expire && rune inject --output ~/.openclaw/workspace/FACTS.md\n\n# Proactive work selection\nNEXT_TASK=$(rune next-task --json)\nif [[ \"$NEXT_TASK\" != \"null\" ]]; then\n # Work on the recommended task\nfi" }, { "title": "Session Hooks", "body": "The skill automatically provides secure session hooks via OpenClaw integrations.\n\nFor manual usage, use the secure session handler:\n\n# Secure session hooks (input sanitized automatically)\n./rune-session-handler.sh start # Loads dynamic context safely\n./rune-session-handler.sh end # Tracks session style safely\n\n# Direct usage (SECURE - input is sanitized):\nSAFE_MESSAGE=$(echo \"$MESSAGE\" | head -c 200 | tr -d '`$(){}[]|;&<>' | sed 's/[^a-zA-Z0-9 ._-]//g')\nrune recall \"$SAFE_MESSAGE\" --limit 10\n\nโš ๏ธ Security Note: Never pass unsanitized user input directly to shell commands. Always use the provided session handler or sanitize input manually." }, { "title": "Architecture", "body": "SQLite Database: All memory stored in ~/.openclaw/memory.db\nLocal LLM Integration: Ollama for relevance scoring and extraction\nCloud API Support: Anthropic, OpenAI for advanced reasoning\nLocal-First Design: Works completely offline with Ollama (cloud APIs optional for advanced features)" }, { "title": "Memory Categories", "body": "person: Information about people (names, roles, preferences)\nproject: Project status, phases, decisions\ntool: How to use tools and their quirks\nlesson: Mistakes to avoid, best practices\ndecision: Why certain choices were made\npreference: User likes/dislikes, settings\nenvironment: System configs, non-sensitive settings (โš ๏ธ NEVER store credentials!)" }, { "title": "Data Storage", "body": "What Rune Stores:\n\nFacts you explicitly add via rune add\nSession interaction patterns for learning (conversation style, not content)\nProject states and task recommendations\n\nWhat Rune Does NOT Store (by default):\n\nFull conversation transcripts (unless you run extract manually)\nAPI keys or credentials (use environment variables instead)\nSensitive personal information (unless you explicitly add it)" }, { "title": "Installation Security", "body": "NPM Dependencies:\n\nInstallation fetches dependencies from npm registry at install time\nReview package.json for dependencies if security is critical\nGlobal npm install runs lifecycle scripts (standard npm behavior)\nConsider installing in isolated environment (container/VM) for high-security use\n\nSession Security:\n\nFixed CVE-2026-0001: Input sanitization prevents shell injection\nAll user input sanitized before shell execution\nSession handler validates and limits input length" }, { "title": "Cloud API Usage", "body": "Optional: Rune can use OpenAI/Anthropic APIs for fact extraction and scoring\nLocal-first: Works completely offline with Ollama (recommended for privacy)\nYour choice: Configure which engines to use in your setup" }, { "title": "Privacy Recommendations", "body": "Use local models (Ollama) for maximum privacy\nAvoid cloud APIs if processing sensitive information\nReview stored facts periodically with rune search\nNever store credentials in memory - use environment variables\n\nPrivacy Best Practices:\n\nNever run rune add with sensitive data (passwords, API keys, personal info)\nUse rune extract carefully - review files before extracting facts\nConfigure Ollama for local-only operation if you want zero cloud usage\nReview your ~/.openclaw/workspace/FACTS.md periodically\n\nInstallation Changes:\n\nAdds memory maintenance commands to HEARTBEAT.md (if present)\nCreates ~/.openclaw/memory.db database file\nSession hooks may process conversation metadata (not full content) for learning" }, { "title": "Performance Metrics", "body": "With Rune, your agent will:\n\nโœ… Remember context between sessions without burning tokens\nโœ… Pick relevant facts dynamically vs dumping everything\nโœ… Get measurably better at avoiding repeated mistakes\nโœ… Work autonomously on projects between check-ins\nโœ… Learn your interaction patterns and adapt responses" }, { "title": "Advanced Configuration", "body": "# Tune relevance scoring\nrune score \"your query\" --threshold 0.6 --model llama3.1:8b\n\n# Configure forgetting curves \nrune forget --decay-rate 0.03 --grace-days 45\n\n# Cross-session pattern analysis\nrune cross-session --days 90 --min-sessions 5" }, { "title": "Automated Maintenance & Performance", "body": "Rune performs best with regular maintenance. Here are automation strategies:" }, { "title": "Cron Job Setup", "body": "Daily Maintenance (3 AM)\n\n# Expire working memory and regenerate context\n0 3 * * * /usr/local/bin/rune expire && /usr/local/bin/rune inject --output ~/.openclaw/workspace/FACTS.md\n\nWeekly Optimization (Sunday 2 AM)\n\n# Consolidate memory and run self-review\n0 2 * * 0 /usr/local/bin/rune consolidate --auto-prioritize && /usr/local/bin/rune self-review --days 7\n\nMonthly Deep Clean (1st of month, 1 AM)\n\n# Pattern analysis and database optimization\n0 1 1 * * /usr/local/bin/rune pattern-analysis --days 30 && sqlite3 ~/.openclaw/memory.db \"VACUUM; ANALYZE;\"" }, { "title": "Database Backup", "body": "# Daily backup at 4 AM\n0 4 * * * cp ~/.openclaw/memory.db ~/.openclaw/memory.db.backup.$(date +\\%Y\\%m\\%d)\n# Keep last 7 days\n5 4 * * * find ~/.openclaw -name \"memory.db.backup.*\" -mtime +7 -delete" }, { "title": "Performance Benefits", "body": "๐Ÿงน Memory stays lean: Auto-removes expired facts\nโšก Faster queries: Regular consolidation prevents bloat\n๐Ÿ“ˆ Self-improvement: Pattern detection catches recurring issues\n๐Ÿ”„ Current context: FACTS.md regenerated with latest data\n๐Ÿ’พ Data protection: Automated backups prevent loss" }, { "title": "Memory Health Monitoring", "body": "# Check database size and fact count\nrune stats\n\n# Review recent patterns\nrune pattern-analysis --days 7\n\n# Check consolidation opportunities \nrune consolidate --dry-run" }, { "title": "Troubleshooting", "body": "Memory growing too large?\n\nRun rune consolidate to merge similar facts\nUse rune forget to apply forgetting curves\nCheck rune stats for database size\n\nRelevance scoring not working?\n\nEnsure Ollama is running: systemctl status ollama\nTest model: rune score \"test\" --engine ollama\nFall back to anthropic/openai engines\n\nContext injection too verbose?\n\nLower relevance threshold: --threshold 0.6\nUse token budgeting: rune budget \"query\" --tokens 300" }, { "title": "Contributing", "body": "Rune is open source. Contributions welcome:\n\nMemory Science: Better consolidation algorithms, forgetting curves\nLLM Integration: New scoring engines, extraction methods\nUI/UX: Better command interfaces, visualization tools\nPerformance: Speed optimizations, memory efficiency" }, { "title": "License", "body": "MIT License - Use freely, modify as needed.\n\nRune: Because your AI should remember like you do." } ], "body": "Rune - Persistent AI Memory System\n\nRune gives your OpenClaw agent persistent, intelligent memory that gets better over time. No more burning tokens on static context files or forgetting important information between sessions.\n\nWhat Rune Does\n๐Ÿง  Smart Memory Management\nDynamic Context Injection: AI selects only relevant facts for each conversation\nAccess Pattern Learning: Frequently used facts get prioritized\nForgetting Curves: Unused facts naturally fade like human memory\nMemory Consolidation: Similar facts get merged, verbose ones compressed\n๐ŸŽฏ Session Intelligence\nInteraction Style Detection: Learns if you prefer brainstorm vs deep-work vs debug modes\nBehavioral Pattern Analysis: Tracks your work patterns and preferences over time\nProactive Memory: Volunteers relevant context unprompted (\"last time you worked on this...\")\n๐Ÿ“‹ Project Autopilot\nSmart Task Recommendations: \"What should I work on next?\" with priority scoring\nBlocker Detection: Identifies stuck projects that need intervention\nProject Health Scoring: 0.0-1.0 health scores based on activity and progress\n๐Ÿ“ข Intelligent Notifications\nPriority Classification: Critical/High/Medium/Low/FYI with context analysis\nSmart Timing: Respects quiet hours, batches low-priority updates\nChannel Routing: DM for urgent, Discord for projects, digest for FYI\n๐Ÿ”„ Self-Improvement Loop\nPattern Detection: \"Forgot to use X skill 3 times\" โ†’ automatic escalation\nPerformance Tracking: Measurable improvement over time\nSkill Usage Analysis: Which skills you use vs neglect\nAbout Rune vs rune\n\nRune is the OpenClaw skill name. rune is the CLI tool name. Think of Rune as the \"skill package\" and rune as the \"command-line interface\" - like how the git skill package provides the git CLI.\n\nSkill name: rune (what you install via ClawHub)\nCLI command: rune (what you run in terminal)\nRepository: https://github.com/TheBobLoblaw/rune\nInstallation Disclosure\n\nโš ๏ธ What This Installation Does:\n\nThe Rune skill installation will:\n\nCreate directories: ~/.openclaw/ and subdirectories\nInstall globally: rune CLI via npm (requires npm dependencies)\nCreate database: SQLite database at ~/.openclaw/memory.db\nModify files: Appends integration lines to existing ~/.openclaw/workspace/HEARTBEAT.md\nAdd session hooks: Automatic memory integration for OpenClaw sessions\n\nBefore installing:\n\nBack up HEARTBEAT.md if it contains important data\nReview package.json dependencies if security is critical\nConsider using local models (Ollama) instead of cloud APIs for privacy\nInstallation\n# Via ClawHub (recommended)\nclawhub install rune\n\n# Manual installation\ngit clone https://github.com/TheBobLoblaw/rune\ncd rune\nnpm install --production\nnpm install -g .\n\nQuick Start\n# Initialize memory system\nrune stats\n\n# Add your first fact\nrune add person cory.name \"Cory - my human user\"\n\n# Generate context for a conversation \nrune context \"Let's work on the website\"\n\n# Get task recommendations\nrune next-task\n\n# Weekly self-review\nrune self-review --days 7\n\nCore Commands\nMemory Management\nrune add - Store a fact\nrune search - Find facts\nrune recall - Smart multi-source recall\nrune inject - Generate context file for agent\nIntelligence Features\nrune context - Dynamic context for message\nrune score - Relevance scoring\nrune proactive - Volunteer relevant context\nrune session-style - Detect interaction style\nProject Management\nrune project-state - Track project phases/blockers\nrune next-task - Smart task picker\nrune stuck-projects - Find blocked work\nAdvanced Features\nrune temporal \"last Tuesday\" - Time-based queries\nrune consolidate - Memory optimization\nrune forget - Apply forgetting curves\nrune pattern-analysis - Detect behavioral patterns\nIntegration with OpenClaw\nHeartbeat Integration\n\nAdd to your HEARTBEAT.md:\n\n# Memory maintenance\nrune expire && rune inject --output ~/.openclaw/workspace/FACTS.md\n\n# Proactive work selection\nNEXT_TASK=$(rune next-task --json)\nif [[ \"$NEXT_TASK\" != \"null\" ]]; then\n # Work on the recommended task\nfi\n\nSession Hooks\n\nThe skill automatically provides secure session hooks via OpenClaw integrations.\n\nFor manual usage, use the secure session handler:\n\n# Secure session hooks (input sanitized automatically)\n./rune-session-handler.sh start # Loads dynamic context safely\n./rune-session-handler.sh end # Tracks session style safely\n\n# Direct usage (SECURE - input is sanitized):\nSAFE_MESSAGE=$(echo \"$MESSAGE\" | head -c 200 | tr -d '`$(){}[]|;&<>' | sed 's/[^a-zA-Z0-9 ._-]//g')\nrune recall \"$SAFE_MESSAGE\" --limit 10\n\n\nโš ๏ธ Security Note: Never pass unsanitized user input directly to shell commands. Always use the provided session handler or sanitize input manually.\n\nArchitecture\nSQLite Database: All memory stored in ~/.openclaw/memory.db\nLocal LLM Integration: Ollama for relevance scoring and extraction\nCloud API Support: Anthropic, OpenAI for advanced reasoning\nLocal-First Design: Works completely offline with Ollama (cloud APIs optional for advanced features)\nMemory Categories\nperson: Information about people (names, roles, preferences)\nproject: Project status, phases, decisions\ntool: How to use tools and their quirks\nlesson: Mistakes to avoid, best practices\ndecision: Why certain choices were made\npreference: User likes/dislikes, settings\nenvironment: System configs, non-sensitive settings (โš ๏ธ NEVER store credentials!)\nโš ๏ธ Security & Privacy\nData Storage\n\nWhat Rune Stores:\n\nFacts you explicitly add via rune add\nSession interaction patterns for learning (conversation style, not content)\nProject states and task recommendations\n\nWhat Rune Does NOT Store (by default):\n\nFull conversation transcripts (unless you run extract manually)\nAPI keys or credentials (use environment variables instead)\nSensitive personal information (unless you explicitly add it)\nInstallation Security\n\nNPM Dependencies:\n\nInstallation fetches dependencies from npm registry at install time\nReview package.json for dependencies if security is critical\nGlobal npm install runs lifecycle scripts (standard npm behavior)\nConsider installing in isolated environment (container/VM) for high-security use\n\nSession Security:\n\nFixed CVE-2026-0001: Input sanitization prevents shell injection\nAll user input sanitized before shell execution\nSession handler validates and limits input length\nCloud API Usage\nOptional: Rune can use OpenAI/Anthropic APIs for fact extraction and scoring\nLocal-first: Works completely offline with Ollama (recommended for privacy)\nYour choice: Configure which engines to use in your setup\nPrivacy Recommendations\nUse local models (Ollama) for maximum privacy\nAvoid cloud APIs if processing sensitive information\nReview stored facts periodically with rune search\nNever store credentials in memory - use environment variables\n\nPrivacy Best Practices:\n\nNever run rune add with sensitive data (passwords, API keys, personal info)\nUse rune extract carefully - review files before extracting facts\nConfigure Ollama for local-only operation if you want zero cloud usage\nReview your ~/.openclaw/workspace/FACTS.md periodically\n\nInstallation Changes:\n\nAdds memory maintenance commands to HEARTBEAT.md (if present)\nCreates ~/.openclaw/memory.db database file\nSession hooks may process conversation metadata (not full content) for learning\nPerformance Metrics\n\nWith Rune, your agent will:\n\nโœ… Remember context between sessions without burning tokens\nโœ… Pick relevant facts dynamically vs dumping everything\nโœ… Get measurably better at avoiding repeated mistakes\nโœ… Work autonomously on projects between check-ins\nโœ… Learn your interaction patterns and adapt responses\nAdvanced Configuration\n# Tune relevance scoring\nrune score \"your query\" --threshold 0.6 --model llama3.1:8b\n\n# Configure forgetting curves \nrune forget --decay-rate 0.03 --grace-days 45\n\n# Cross-session pattern analysis\nrune cross-session --days 90 --min-sessions 5\n\nAutomated Maintenance & Performance\n\nRune performs best with regular maintenance. Here are automation strategies:\n\nCron Job Setup\n\nDaily Maintenance (3 AM)\n\n# Expire working memory and regenerate context\n0 3 * * * /usr/local/bin/rune expire && /usr/local/bin/rune inject --output ~/.openclaw/workspace/FACTS.md\n\n\nWeekly Optimization (Sunday 2 AM)\n\n# Consolidate memory and run self-review\n0 2 * * 0 /usr/local/bin/rune consolidate --auto-prioritize && /usr/local/bin/rune self-review --days 7\n\n\nMonthly Deep Clean (1st of month, 1 AM)\n\n# Pattern analysis and database optimization\n0 1 1 * * /usr/local/bin/rune pattern-analysis --days 30 && sqlite3 ~/.openclaw/memory.db \"VACUUM; ANALYZE;\"\n\nDatabase Backup\n# Daily backup at 4 AM\n0 4 * * * cp ~/.openclaw/memory.db ~/.openclaw/memory.db.backup.$(date +\\%Y\\%m\\%d)\n# Keep last 7 days\n5 4 * * * find ~/.openclaw -name \"memory.db.backup.*\" -mtime +7 -delete\n\nPerformance Benefits\n๐Ÿงน Memory stays lean: Auto-removes expired facts\nโšก Faster queries: Regular consolidation prevents bloat\n๐Ÿ“ˆ Self-improvement: Pattern detection catches recurring issues\n๐Ÿ”„ Current context: FACTS.md regenerated with latest data\n๐Ÿ’พ Data protection: Automated backups prevent loss\nMemory Health Monitoring\n# Check database size and fact count\nrune stats\n\n# Review recent patterns\nrune pattern-analysis --days 7\n\n# Check consolidation opportunities \nrune consolidate --dry-run\n\nTroubleshooting\n\nMemory growing too large?\n\nRun rune consolidate to merge similar facts\nUse rune forget to apply forgetting curves\nCheck rune stats for database size\n\nRelevance scoring not working?\n\nEnsure Ollama is running: systemctl status ollama\nTest model: rune score \"test\" --engine ollama\nFall back to anthropic/openai engines\n\nContext injection too verbose?\n\nLower relevance threshold: --threshold 0.6\nUse token budgeting: rune budget \"query\" --tokens 300\nContributing\n\nRune is open source. Contributions welcome:\n\nMemory Science: Better consolidation algorithms, forgetting curves\nLLM Integration: New scoring engines, extraction methods\nUI/UX: Better command interfaces, visualization tools\nPerformance: Speed optimizations, memory efficiency\nLicense\n\nMIT License - Use freely, modify as needed.\n\nRune: Because your AI should remember like you do." }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/TheBobLoblaw/rune", "publisherUrl": "https://clawhub.ai/TheBobLoblaw/rune", "owner": "TheBobLoblaw", "version": "1.1.5", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/rune", "downloadUrl": "https://openagent3.xyz/downloads/rune", "agentUrl": "https://openagent3.xyz/skills/rune/agent", "manifestUrl": "https://openagent3.xyz/skills/rune/agent.json", "briefUrl": "https://openagent3.xyz/skills/rune/agent.md" } }