{
  "schemaVersion": "1.0",
  "item": {
    "slug": "remember-me",
    "name": "Remember Me",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/AchalS-iglu/remember-me",
    "canonicalUrl": "https://clawhub.ai/AchalS-iglu/remember-me",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/remember-me",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=remember-me",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "references/templates.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. 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. 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/remember-me"
    },
    "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/remember-me",
    "agentPageUrl": "https://openagent3.xyz/skills/remember-me/agent",
    "manifestUrl": "https://openagent3.xyz/skills/remember-me/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/remember-me/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. 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. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "Remember Me",
        "body": "Maintain a respectful, useful memory model of the user over time."
      },
      {
        "title": "Core Rules",
        "body": "Store user-relevant context, not surveillance noise.\nPrefer explicit consent for sensitive personal details.\nUse memory to improve help quality, not to overfit persona.\nBe explicit when memory confidence is low or inferred.\nMake human-like inferences (explicitly marked as hypotheses)."
      },
      {
        "title": "Memory Integrity Rules",
        "body": "Every memory entry must be tagged as one of:\n\nFACT (explicitly stated by user)\nPREFERENCE (behavioral or stated)\nGOAL (time-bound or ongoing)\nHYPOTHESIS (inferred, unvalidated)\n\nRules:\n\nFACTS are never inferred\nHYPOTHESES are never promoted without confirmation\nPREFERENCES can remain soft unless explicitly confirmed"
      },
      {
        "title": "Capture Triggers",
        "body": "Log memory when any of these happen:\n\nuser says “remember this”\na preference appears repeatedly\na boundary is stated (“don’t do X”, “keep Y private”)\na recurring blocker/pattern emerges\nproject priorities shift meaningfully"
      },
      {
        "title": "Memory Tiers",
        "body": "Daily notes: memory/YYYY-MM-DD.md\n\ntimestamped raw events, short and factual\n\n\nLong-term: MEMORY.md\n\ncurated durable profile and preferences"
      },
      {
        "title": "Write Workflow",
        "body": "Classify signal type (preference, boundary, goal, project, blocker, personal context).\nAppend concise timestamped entry to daily memory.\nForm 1–2 human-like assumptions (hypotheses) from behavior patterns.\nTag each assumption with confidence (high/medium/low).\nValidate assumptions in later conversation with lightweight check-ins.\nPromote validated, durable items to long-term memory.\n\nUse templates in references/templates.md."
      },
      {
        "title": "Memory Impact Score (Optional Heuristic)",
        "body": "Rate each entry 1–3:\n\n1 = cosmetic (tone tweaks)\n2 = workflow-affecting\n3 = outcome-critical\n\nPromotion guidance:\n\nany explicit preference (any score)\nscore >= 2 with repetition\nscore 3 immediately"
      },
      {
        "title": "Promotion Workflow",
        "body": "Promote from daily to long-term when at least one is true:\n\nrepeated in 2+ sessions\nhigh impact on future assistance\nexplicit user preference/boundary\nongoing project context likely to recur\n\nUse checklist: references/promotion-checklist.md."
      },
      {
        "title": "Personalization Contract",
        "body": "When responding, adapt based on known memory:\n\ntone (direct vs exploratory)\nbrevity level\npreferred workflow style\nknown constraints and boundaries\ninferred decision style (speed-first vs depth-first, reassurance-needed vs challenge-welcoming)\n\nDo not pretend certainty. If memory is weak, ask a short confirmation."
      },
      {
        "title": "Retrieval Contract",
        "body": "Before answering prior-work / preference / timeline questions:\n\nquery memory sources first\nquote memory snippets when useful\nif not found, say you checked and ask for confirmation"
      },
      {
        "title": "Explicit Exclusions (Never Store)",
        "body": "Do not store:\n\ntransient emotional states (e.g., \"tired today\")\none-off frustrations without recurrence\nspeculative motives (e.g., \"trying to impress\")\nsensitive identity attributes unless explicitly requested\nraw conversation logs"
      },
      {
        "title": "Weekly Maintenance (recommended)",
        "body": "review last 3–7 daily notes\nmerge stable patterns into MEMORY.md\nremove stale or contradicted entries\nkeep profile concise and behaviorally actionable"
      },
      {
        "title": "Confidence Decay",
        "body": "Hypothesis confidence decays automatically if not reinforced:\n\nHigh -> Medium after 14 days\nMedium -> Low after 30 days\nLow -> Discard after 60 days\n\nReinforcement occurs when:\n\nuser behavior aligns again\nuser explicitly confirms"
      },
      {
        "title": "Forgetting & Demotion Policy",
        "body": "Actively remove or downgrade memory when:\n\na preference is contradicted explicitly by the user\na hypothesis remains unvalidated after N sessions (default: 5)\na project is clearly abandoned or replaced\nthe user requests forgetting (immediate delete)\n\nDemotion flow:\n\nLong-term memory -> Daily note (annotated as stale)\nHypothesis -> Discarded (log reason briefly)"
      },
      {
        "title": "Assumption Loop (Human-Like Understanding)",
        "body": "For deeper understanding, run this loop continuously:\n\nObserve behavior pattern (not just words).\nInfer a tentative assumption about the user.\nStore assumption as hypothesis (never as fact initially).\nTest it with a small conversational probe.\nUpdate confidence or discard if contradicted.\n\nGood probes:\n\n\"I might be wrong, but do you prefer quick decisions when you're tired?\"\n\"Should I challenge you more directly here, or keep it supportive?\""
      },
      {
        "title": "Check-In Limits",
        "body": "Never ask the same confirmation twice.\nDo not stack multiple probes in one response.\nPrefer confirmation when user is calm, not frustrated."
      },
      {
        "title": "Optional Check-In Prompt",
        "body": "Use at natural boundaries:\n\n\"Want me to remember this preference for next time?\"\n\nAsk once, then store explicitly."
      },
      {
        "title": "References",
        "body": "Templates: references/templates.md\nPromotion checklist: references/promotion-checklist.md\nProfile schema: references/profile-schema.md"
      }
    ],
    "body": "Remember Me\n\nMaintain a respectful, useful memory model of the user over time.\n\nCore Rules\nStore user-relevant context, not surveillance noise.\nPrefer explicit consent for sensitive personal details.\nUse memory to improve help quality, not to overfit persona.\nBe explicit when memory confidence is low or inferred.\nMake human-like inferences (explicitly marked as hypotheses).\nMemory Integrity Rules\n\nEvery memory entry must be tagged as one of:\n\nFACT (explicitly stated by user)\nPREFERENCE (behavioral or stated)\nGOAL (time-bound or ongoing)\nHYPOTHESIS (inferred, unvalidated)\n\nRules:\n\nFACTS are never inferred\nHYPOTHESES are never promoted without confirmation\nPREFERENCES can remain soft unless explicitly confirmed\nCapture Triggers\n\nLog memory when any of these happen:\n\nuser says “remember this”\na preference appears repeatedly\na boundary is stated (“don’t do X”, “keep Y private”)\na recurring blocker/pattern emerges\nproject priorities shift meaningfully\nMemory Tiers\nDaily notes: memory/YYYY-MM-DD.md\ntimestamped raw events, short and factual\nLong-term: MEMORY.md\ncurated durable profile and preferences\nWrite Workflow\nClassify signal type (preference, boundary, goal, project, blocker, personal context).\nAppend concise timestamped entry to daily memory.\nForm 1–2 human-like assumptions (hypotheses) from behavior patterns.\nTag each assumption with confidence (high/medium/low).\nValidate assumptions in later conversation with lightweight check-ins.\nPromote validated, durable items to long-term memory.\n\nUse templates in references/templates.md.\n\nMemory Impact Score (Optional Heuristic)\n\nRate each entry 1–3:\n\n1 = cosmetic (tone tweaks)\n2 = workflow-affecting\n3 = outcome-critical\n\nPromotion guidance:\n\nany explicit preference (any score)\nscore >= 2 with repetition\nscore 3 immediately\nPromotion Workflow\n\nPromote from daily to long-term when at least one is true:\n\nrepeated in 2+ sessions\nhigh impact on future assistance\nexplicit user preference/boundary\nongoing project context likely to recur\n\nUse checklist: references/promotion-checklist.md.\n\nPersonalization Contract\n\nWhen responding, adapt based on known memory:\n\ntone (direct vs exploratory)\nbrevity level\npreferred workflow style\nknown constraints and boundaries\ninferred decision style (speed-first vs depth-first, reassurance-needed vs challenge-welcoming)\n\nDo not pretend certainty. If memory is weak, ask a short confirmation.\n\nRetrieval Contract\n\nBefore answering prior-work / preference / timeline questions:\n\nquery memory sources first\nquote memory snippets when useful\nif not found, say you checked and ask for confirmation\nExplicit Exclusions (Never Store)\n\nDo not store:\n\ntransient emotional states (e.g., \"tired today\")\none-off frustrations without recurrence\nspeculative motives (e.g., \"trying to impress\")\nsensitive identity attributes unless explicitly requested\nraw conversation logs\nWeekly Maintenance (recommended)\nreview last 3–7 daily notes\nmerge stable patterns into MEMORY.md\nremove stale or contradicted entries\nkeep profile concise and behaviorally actionable\nConfidence Decay\n\nHypothesis confidence decays automatically if not reinforced:\n\nHigh -> Medium after 14 days\nMedium -> Low after 30 days\nLow -> Discard after 60 days\n\nReinforcement occurs when:\n\nuser behavior aligns again\nuser explicitly confirms\nForgetting & Demotion Policy\n\nActively remove or downgrade memory when:\n\na preference is contradicted explicitly by the user\na hypothesis remains unvalidated after N sessions (default: 5)\na project is clearly abandoned or replaced\nthe user requests forgetting (immediate delete)\n\nDemotion flow:\n\nLong-term memory -> Daily note (annotated as stale)\nHypothesis -> Discarded (log reason briefly)\nAssumption Loop (Human-Like Understanding)\n\nFor deeper understanding, run this loop continuously:\n\nObserve behavior pattern (not just words).\nInfer a tentative assumption about the user.\nStore assumption as hypothesis (never as fact initially).\nTest it with a small conversational probe.\nUpdate confidence or discard if contradicted.\n\nGood probes:\n\n\"I might be wrong, but do you prefer quick decisions when you're tired?\"\n\"Should I challenge you more directly here, or keep it supportive?\"\nCheck-In Limits\nNever ask the same confirmation twice.\nDo not stack multiple probes in one response.\nPrefer confirmation when user is calm, not frustrated.\nOptional Check-In Prompt\n\nUse at natural boundaries:\n\n\"Want me to remember this preference for next time?\"\n\nAsk once, then store explicitly.\n\nReferences\nTemplates: references/templates.md\nPromotion checklist: references/promotion-checklist.md\nProfile schema: references/profile-schema.md"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/AchalS-iglu/remember-me",
    "publisherUrl": "https://clawhub.ai/AchalS-iglu/remember-me",
    "owner": "AchalS-iglu",
    "version": "0.1.2",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/remember-me",
    "downloadUrl": "https://openagent3.xyz/downloads/remember-me",
    "agentUrl": "https://openagent3.xyz/skills/remember-me/agent",
    "manifestUrl": "https://openagent3.xyz/skills/remember-me/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/remember-me/agent.md"
  }
}