{
  "schemaVersion": "1.0",
  "item": {
    "slug": "correction-memory",
    "name": "Correction Memory",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/donovanpankratz-del/correction-memory",
    "canonicalUrl": "https://clawhub.ai/donovanpankratz-del/correction-memory",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/correction-memory",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=correction-memory",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "references/correction-tracker-template.js"
    ],
    "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",
      "slug": "correction-memory",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-30T20:35:58.824Z",
      "expiresAt": "2026-05-07T20:35:58.824Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=correction-memory",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=correction-memory",
        "contentDisposition": "attachment; filename=\"correction-memory-1.1.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "correction-memory"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/correction-memory"
    },
    "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/correction-memory",
    "agentPageUrl": "https://openagent3.xyz/skills/correction-memory/agent",
    "manifestUrl": "https://openagent3.xyz/skills/correction-memory/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/correction-memory/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": "The Problem",
        "body": "When you correct an agent, that correction evaporates after the session. Next time you spawn the same agent type, it makes the same mistake. There's no memory of what you've already taught it."
      },
      {
        "title": "What This Skill Installs",
        "body": "lib/correction-tracker.js — logs corrections per agent type to memory/corrections/[AgentType].jsonl\nHook into agent-context-loader.js — correction preamble prepended to spawns automatically (if intent-engineering is also installed)"
      },
      {
        "title": "Step 1 — Install correction-tracker",
        "body": "cp references/correction-tracker-template.js $OPENCLAW_WORKSPACE/lib/correction-tracker.js\n\nVerify it runs:\n\nnode $OPENCLAW_WORKSPACE/lib/correction-tracker.js"
      },
      {
        "title": "Step 2 — Wire agent-context-loader (if using intent-engineering)",
        "body": "If lib/agent-context-loader.js is installed (from intent-engineering skill), correction injection is automatic — no wiring needed. The loader checks for correction-tracker.js at startup and loads it if present.\n\nIf you are NOT using intent-engineering, add this to your spawn logic manually:\n\nconst { buildCorrectionPreamble } = require('./lib/correction-tracker');\n\nconst agentType   = 'CoderAgent'; // or whatever agent you're spawning\nconst corrections = buildCorrectionPreamble(agentType, workspaceRoot);\nconst fullTask    = corrections ? corrections + '\\n\\n---\\n\\n' + originalTask : originalTask;"
      },
      {
        "title": "Programmatic",
        "body": "const { logCorrection } = require('./lib/correction-tracker');\n\nlogCorrection(\n  'CoderAgent',                                    // agent type\n  'Used ESM import instead of require()',          // what was wrong\n  'Always use require() for Node.js stdlib modules', // correct behavior\n  workspaceRoot,\n  { session_channel: 'discord' }                  // optional metadata\n);"
      },
      {
        "title": "Via main agent (natural language)",
        "body": "Just tell the main agent:\n\n\"Note that [AgentType]: [what it did wrong] — [correct behavior]\"\n\nThe main agent will log it programmatically."
      },
      {
        "title": "How Corrections Are Replayed",
        "body": "On every subagent spawn, agent-context-loader detects the agent type from the task description and prepends:\n\n## Corrections from Previous Sessions\n\nThe following corrections were logged for CoderAgent. Apply these behaviors:\n\n1. **[2026-03-01] Issue:** Used ESM import instead of require()\n   **Correction:** Always use require() for Node.js stdlib modules\n\nOnly corrections from the last 30 days are injected. Older corrections expire automatically — stale rules don't accumulate."
      },
      {
        "title": "Viewing Corrections",
        "body": "# All corrections for an agent type\ncat $OPENCLAW_WORKSPACE/memory/corrections/CoderAgent.jsonl | jq .\n\n# List all agent types with corrections\nls $OPENCLAW_WORKSPACE/memory/corrections/\n\n# Count corrections per agent\nfor f in $OPENCLAW_WORKSPACE/memory/corrections/*.jsonl; do\n  echo \"$(basename $f .jsonl): $(wc -l < $f) corrections\"\ndone"
      },
      {
        "title": "Agent Type Detection",
        "body": "The loader auto-detects agent type from the task description. Default rules:\n\nTask keywordsAgent typecode, coder, impl, debugCoderAgentwrit, author, novel, chapterAuthorAgentworld, buildWorldbuilderAgent(anything else)general\n\nTo add custom agent types, edit detectAgentType() in agent-context-loader.js."
      },
      {
        "title": "References",
        "body": "references/correction-tracker-template.js — Full implementation of correction-tracker.js"
      }
    ],
    "body": "Correction Memory\nThe Problem\n\nWhen you correct an agent, that correction evaporates after the session. Next time you spawn the same agent type, it makes the same mistake. There's no memory of what you've already taught it.\n\nWhat This Skill Installs\nlib/correction-tracker.js — logs corrections per agent type to memory/corrections/[AgentType].jsonl\nHook into agent-context-loader.js — correction preamble prepended to spawns automatically (if intent-engineering is also installed)\nInstallation\nStep 1 — Install correction-tracker\ncp references/correction-tracker-template.js $OPENCLAW_WORKSPACE/lib/correction-tracker.js\n\n\nVerify it runs:\n\nnode $OPENCLAW_WORKSPACE/lib/correction-tracker.js\n\nStep 2 — Wire agent-context-loader (if using intent-engineering)\n\nIf lib/agent-context-loader.js is installed (from intent-engineering skill), correction injection is automatic — no wiring needed. The loader checks for correction-tracker.js at startup and loads it if present.\n\nIf you are NOT using intent-engineering, add this to your spawn logic manually:\n\nconst { buildCorrectionPreamble } = require('./lib/correction-tracker');\n\nconst agentType   = 'CoderAgent'; // or whatever agent you're spawning\nconst corrections = buildCorrectionPreamble(agentType, workspaceRoot);\nconst fullTask    = corrections ? corrections + '\\n\\n---\\n\\n' + originalTask : originalTask;\n\nLogging Corrections\nProgrammatic\nconst { logCorrection } = require('./lib/correction-tracker');\n\nlogCorrection(\n  'CoderAgent',                                    // agent type\n  'Used ESM import instead of require()',          // what was wrong\n  'Always use require() for Node.js stdlib modules', // correct behavior\n  workspaceRoot,\n  { session_channel: 'discord' }                  // optional metadata\n);\n\nVia main agent (natural language)\n\nJust tell the main agent:\n\n\"Note that [AgentType]: [what it did wrong] — [correct behavior]\"\n\nThe main agent will log it programmatically.\n\nHow Corrections Are Replayed\n\nOn every subagent spawn, agent-context-loader detects the agent type from the task description and prepends:\n\n## Corrections from Previous Sessions\n\nThe following corrections were logged for CoderAgent. Apply these behaviors:\n\n1. **[2026-03-01] Issue:** Used ESM import instead of require()\n   **Correction:** Always use require() for Node.js stdlib modules\n\n\nOnly corrections from the last 30 days are injected. Older corrections expire automatically — stale rules don't accumulate.\n\nViewing Corrections\n# All corrections for an agent type\ncat $OPENCLAW_WORKSPACE/memory/corrections/CoderAgent.jsonl | jq .\n\n# List all agent types with corrections\nls $OPENCLAW_WORKSPACE/memory/corrections/\n\n# Count corrections per agent\nfor f in $OPENCLAW_WORKSPACE/memory/corrections/*.jsonl; do\n  echo \"$(basename $f .jsonl): $(wc -l < $f) corrections\"\ndone\n\nAgent Type Detection\n\nThe loader auto-detects agent type from the task description. Default rules:\n\nTask keywords\tAgent type\ncode, coder, impl, debug\tCoderAgent\nwrit, author, novel, chapter\tAuthorAgent\nworld, build\tWorldbuilderAgent\n(anything else)\tgeneral\n\nTo add custom agent types, edit detectAgentType() in agent-context-loader.js.\n\nReferences\nreferences/correction-tracker-template.js — Full implementation of correction-tracker.js"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/donovanpankratz-del/correction-memory",
    "publisherUrl": "https://clawhub.ai/donovanpankratz-del/correction-memory",
    "owner": "donovanpankratz-del",
    "version": "1.1.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/correction-memory",
    "downloadUrl": "https://openagent3.xyz/downloads/correction-memory",
    "agentUrl": "https://openagent3.xyz/skills/correction-memory/agent",
    "manifestUrl": "https://openagent3.xyz/skills/correction-memory/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/correction-memory/agent.md"
  }
}