{
  "schemaVersion": "1.0",
  "item": {
    "slug": "auto-context",
    "name": "Auto Context",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/wpank/auto-context",
    "canonicalUrl": "https://clawhub.ai/wpank/auto-context",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/auto-context",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=auto-context",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "README.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": "2026-04-23T16:43:11.935Z",
      "expiresAt": "2026-04-30T16:43:11.935Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
        "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.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/auto-context"
    },
    "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/auto-context",
    "agentPageUrl": "https://openagent3.xyz/skills/auto-context/agent",
    "manifestUrl": "https://openagent3.xyz/skills/auto-context/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/auto-context/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": "Auto-Context — Situational Awareness Protocol (Meta-Skill)",
        "body": "Before you act, understand where you are. This skill ensures the AI loads critical project context automatically before any major action, preventing wasted effort, duplicate work, and misaligned implementations."
      },
      {
        "title": "OpenClaw / Moltbot / Clawbot",
        "body": "npx clawhub@latest install auto-context"
      },
      {
        "title": "When to Activate",
        "body": "This skill triggers automatically based on the current action. Do not wait for the user to ask — proactively load context when any of these conditions are met.\n\nTriggerWhyMinimum ContextStarting a new taskUnderstand priorities, avoid conflictsCritical + HighImplementing a featureKnow the plan, constraints, recent changesCritical + HighRefactoring codeUnderstand what changed recently, what's plannedCritical + High + MediumDebugging an issueCheck recent changes, known issues, discoveriesCritical + High + MediumPlanning or scoping workFull picture of roadmap, backlog, progressAll levelsSession start or resumeRebuild mental model from last session stateCritical + HighBefore a handoffEnsure nothing is missed in transitionAll levels"
      },
      {
        "title": "Context Files to Read",
        "body": "Read these files in priority order. Stop early if the task is narrow and lower-priority files are clearly irrelevant.\n\nPriorityFilePurposeRead WhenCriticalTODO.mdCurrent tasks, backlog, and prioritiesAlwaysCriticalroadmap.mdPhase status, milestones, project directionAlwaysHightask_plan.mdActive task breakdown and implementation planFile existsHigh.cursor/handoffs/*.mdRecent handoff notes (read last 3 by date)File existsMediumfindings.mdResearch results, discoveries, decisions madeRelevant to taskMediumCHANGELOG.mdRecent changes and their rationaleRelevant to taskLow.cursor/sessions/*.mdSession summaries (read last 2 by date)Planning or debugging"
      },
      {
        "title": "Alternate Locations",
        "body": "Some projects use different paths. Check these fallbacks if primary paths are empty:\n\nPrimaryFallbackTODO.mddocs/TODO.md, ai/TODO.mdroadmap.mddocs/roadmap.md, ROADMAP.mdtask_plan.mddocs/task_plan.md, .cursor/task_plan.mdfindings.mddocs/findings.md, .cursor/findings.md"
      },
      {
        "title": "Step 1: Load Critical Files (Always)",
        "body": "Read TODO.md → Extract: current task, next priorities, blockers\nRead roadmap.md → Extract: current phase, active milestone, upcoming deadlines\n\nIf either critical file is missing, warn the user:\n\n\"No TODO.md found. Consider creating one to track tasks.\""
      },
      {
        "title": "Step 2: Load High-Priority Files (If They Exist)",
        "body": "Read task_plan.md → Extract: implementation steps, acceptance criteria\nGlob .cursor/handoffs/*.md → Read last 3 by modification date"
      },
      {
        "title": "Step 3: Load Medium/Low Files (If Relevant)",
        "body": "Only read these when the current task benefits from historical context:\n\nDebugging? — Read findings.md and CHANGELOG.md\nPlanning? — Read everything including session files\nQuick fix? — Skip medium and low entirely"
      },
      {
        "title": "Step 4: Synthesize and Present",
        "body": "After loading, produce a context summary (see format below) before proceeding with the task."
      },
      {
        "title": "Staleness Detection",
        "body": "Check modification dates on all loaded files. Flag files that may contain outdated information.\n\nAgeStatusAction< 24 hoursFreshUse as-is1-7 daysCurrentUse as-is, note the age7-30 daysStaleWarn: \"{file} last updated {N} days ago — verify before relying on it\"> 30 daysOutdatedWarn: \"{file} is {N} days old and may no longer reflect project state\"\n\nTo check file ages on macOS:\n\nstat -f \"%m %N\" TODO.md roadmap.md task_plan.md findings.md CHANGELOG.md 2>/dev/null\n\nOn Linux:\n\nstat -c \"%Y %n\" TODO.md roadmap.md task_plan.md findings.md CHANGELOG.md 2>/dev/null"
      },
      {
        "title": "Context Summary Format",
        "body": "After loading context, present a concise summary using this template. Keep it tight — the goal is awareness, not repetition.\n\n## Context Loaded\n\n**Current Phase:** {phase from roadmap}\n**Active Milestone:** {milestone and progress}\n\n**Current Task:** {from TODO.md or task_plan.md}\n- Status: {in-progress / blocked / not started}\n- Blockers: {any blockers, or \"none\"}\n\n**Recent Changes:**\n- {last 2-3 items from CHANGELOG or handoffs}\n\n**Relevant Findings:**\n- {key discoveries that affect the current task, or \"none\"}\n\n**Stale Warnings:**\n- {any staleness warnings, or \"all context is fresh\"}\n\nIf no context files exist at all, output:\n\n## Context Loaded\n\nNo project context files found. Operating without historical context.\nConsider creating TODO.md and roadmap.md to enable context-aware assistance."
      },
      {
        "title": "Integration Points",
        "body": "This skill connects to other workflow commands and should run as a precursor.\n\nCommandHow Auto-Context Integrates/start-taskLoads full context before beginning work; populates task plan/intentReads roadmap and TODO to validate intent against project direction/workflowProvides the \"understand\" phase of any workflow automatically/progressUses TODO.md and task_plan.md to assess completion status/handoff-and-resumeReads last handoff to rebuild state on resume/session-summaryCross-references loaded context with session actions for accuracy"
      },
      {
        "title": "Execution Order",
        "body": "User triggers action\n  → Auto-Context activates (this skill)\n    → Context summary presented\n      → Primary skill/command executes with full awareness"
      },
      {
        "title": "Quick Reference",
        "body": "Context Loading Checklist:\n  1. Read TODO.md (critical)\n  2. Read roadmap.md (critical)\n  3. Read task_plan.md (if exists)\n  4. Read last 3 handoffs (if exist)\n  5. Check file staleness\n  6. Read findings/changelog (if relevant)\n  7. Present context summary\n  8. Proceed with task"
      },
      {
        "title": "NEVER Do",
        "body": "NEVER skip critical files to save time — leads to duplicate work and conflicting implementations\nNEVER load every file regardless of task — wastes tokens and dilutes focus with irrelevant info\nNEVER ignore staleness warnings — stale context causes decisions based on outdated assumptions\nNEVER read files without summarizing — raw file dumps overwhelm; always synthesize first\nNEVER assume context from memory alone — sessions are stateless; always re-read files\nNEVER silently proceed when no context exists — user should know they're operating blind\nNEVER read handoffs/sessions beyond the limit — last 3 handoffs and last 2 sessions are sufficient"
      }
    ],
    "body": "Auto-Context — Situational Awareness Protocol (Meta-Skill)\n\nBefore you act, understand where you are. This skill ensures the AI loads critical project context automatically before any major action, preventing wasted effort, duplicate work, and misaligned implementations.\n\nInstallation\nOpenClaw / Moltbot / Clawbot\nnpx clawhub@latest install auto-context\n\nWhen to Activate\n\nThis skill triggers automatically based on the current action. Do not wait for the user to ask — proactively load context when any of these conditions are met.\n\nTrigger\tWhy\tMinimum Context\nStarting a new task\tUnderstand priorities, avoid conflicts\tCritical + High\nImplementing a feature\tKnow the plan, constraints, recent changes\tCritical + High\nRefactoring code\tUnderstand what changed recently, what's planned\tCritical + High + Medium\nDebugging an issue\tCheck recent changes, known issues, discoveries\tCritical + High + Medium\nPlanning or scoping work\tFull picture of roadmap, backlog, progress\tAll levels\nSession start or resume\tRebuild mental model from last session state\tCritical + High\nBefore a handoff\tEnsure nothing is missed in transition\tAll levels\nContext Files to Read\n\nRead these files in priority order. Stop early if the task is narrow and lower-priority files are clearly irrelevant.\n\nPriority\tFile\tPurpose\tRead When\nCritical\tTODO.md\tCurrent tasks, backlog, and priorities\tAlways\nCritical\troadmap.md\tPhase status, milestones, project direction\tAlways\nHigh\ttask_plan.md\tActive task breakdown and implementation plan\tFile exists\nHigh\t.cursor/handoffs/*.md\tRecent handoff notes (read last 3 by date)\tFile exists\nMedium\tfindings.md\tResearch results, discoveries, decisions made\tRelevant to task\nMedium\tCHANGELOG.md\tRecent changes and their rationale\tRelevant to task\nLow\t.cursor/sessions/*.md\tSession summaries (read last 2 by date)\tPlanning or debugging\nAlternate Locations\n\nSome projects use different paths. Check these fallbacks if primary paths are empty:\n\nPrimary\tFallback\nTODO.md\tdocs/TODO.md, ai/TODO.md\nroadmap.md\tdocs/roadmap.md, ROADMAP.md\ntask_plan.md\tdocs/task_plan.md, .cursor/task_plan.md\nfindings.md\tdocs/findings.md, .cursor/findings.md\nContext Loading Strategy\nStep 1: Load Critical Files (Always)\nRead TODO.md → Extract: current task, next priorities, blockers\nRead roadmap.md → Extract: current phase, active milestone, upcoming deadlines\n\n\nIf either critical file is missing, warn the user:\n\n\"No TODO.md found. Consider creating one to track tasks.\"\n\nStep 2: Load High-Priority Files (If They Exist)\nRead task_plan.md → Extract: implementation steps, acceptance criteria\nGlob .cursor/handoffs/*.md → Read last 3 by modification date\n\nStep 3: Load Medium/Low Files (If Relevant)\n\nOnly read these when the current task benefits from historical context:\n\nDebugging? — Read findings.md and CHANGELOG.md\nPlanning? — Read everything including session files\nQuick fix? — Skip medium and low entirely\nStep 4: Synthesize and Present\n\nAfter loading, produce a context summary (see format below) before proceeding with the task.\n\nStaleness Detection\n\nCheck modification dates on all loaded files. Flag files that may contain outdated information.\n\nAge\tStatus\tAction\n< 24 hours\tFresh\tUse as-is\n1-7 days\tCurrent\tUse as-is, note the age\n7-30 days\tStale\tWarn: \"{file} last updated {N} days ago — verify before relying on it\"\n> 30 days\tOutdated\tWarn: \"{file} is {N} days old and may no longer reflect project state\"\n\nTo check file ages on macOS:\n\nstat -f \"%m %N\" TODO.md roadmap.md task_plan.md findings.md CHANGELOG.md 2>/dev/null\n\n\nOn Linux:\n\nstat -c \"%Y %n\" TODO.md roadmap.md task_plan.md findings.md CHANGELOG.md 2>/dev/null\n\nContext Summary Format\n\nAfter loading context, present a concise summary using this template. Keep it tight — the goal is awareness, not repetition.\n\n## Context Loaded\n\n**Current Phase:** {phase from roadmap}\n**Active Milestone:** {milestone and progress}\n\n**Current Task:** {from TODO.md or task_plan.md}\n- Status: {in-progress / blocked / not started}\n- Blockers: {any blockers, or \"none\"}\n\n**Recent Changes:**\n- {last 2-3 items from CHANGELOG or handoffs}\n\n**Relevant Findings:**\n- {key discoveries that affect the current task, or \"none\"}\n\n**Stale Warnings:**\n- {any staleness warnings, or \"all context is fresh\"}\n\n\nIf no context files exist at all, output:\n\n## Context Loaded\n\nNo project context files found. Operating without historical context.\nConsider creating TODO.md and roadmap.md to enable context-aware assistance.\n\nIntegration Points\n\nThis skill connects to other workflow commands and should run as a precursor.\n\nCommand\tHow Auto-Context Integrates\n/start-task\tLoads full context before beginning work; populates task plan\n/intent\tReads roadmap and TODO to validate intent against project direction\n/workflow\tProvides the \"understand\" phase of any workflow automatically\n/progress\tUses TODO.md and task_plan.md to assess completion status\n/handoff-and-resume\tReads last handoff to rebuild state on resume\n/session-summary\tCross-references loaded context with session actions for accuracy\nExecution Order\nUser triggers action\n  → Auto-Context activates (this skill)\n    → Context summary presented\n      → Primary skill/command executes with full awareness\n\nQuick Reference\nContext Loading Checklist:\n  1. Read TODO.md (critical)\n  2. Read roadmap.md (critical)\n  3. Read task_plan.md (if exists)\n  4. Read last 3 handoffs (if exist)\n  5. Check file staleness\n  6. Read findings/changelog (if relevant)\n  7. Present context summary\n  8. Proceed with task\n\nNEVER Do\nNEVER skip critical files to save time — leads to duplicate work and conflicting implementations\nNEVER load every file regardless of task — wastes tokens and dilutes focus with irrelevant info\nNEVER ignore staleness warnings — stale context causes decisions based on outdated assumptions\nNEVER read files without summarizing — raw file dumps overwhelm; always synthesize first\nNEVER assume context from memory alone — sessions are stateless; always re-read files\nNEVER silently proceed when no context exists — user should know they're operating blind\nNEVER read handoffs/sessions beyond the limit — last 3 handoffs and last 2 sessions are sufficient"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/wpank/auto-context",
    "publisherUrl": "https://clawhub.ai/wpank/auto-context",
    "owner": "wpank",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/auto-context",
    "downloadUrl": "https://openagent3.xyz/downloads/auto-context",
    "agentUrl": "https://openagent3.xyz/skills/auto-context/agent",
    "manifestUrl": "https://openagent3.xyz/skills/auto-context/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/auto-context/agent.md"
  }
}