{ "schemaVersion": "1.0", "item": { "slug": "reclaw", "name": "Reclaw", "source": "tencent", "type": "skill", "category": "通讯协作", "sourceUrl": "https://clawhub.ai/maxpetretta/reclaw", "canonicalUrl": "https://clawhub.ai/maxpetretta/reclaw", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/reclaw", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=reclaw", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "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/reclaw" }, "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/reclaw", "agentPageUrl": "https://openagent3.xyz/skills/reclaw/agent", "manifestUrl": "https://openagent3.xyz/skills/reclaw/agent.json", "briefUrl": "https://openagent3.xyz/skills/reclaw/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": "Reclaw Memory System", "body": "Reclaw is an append-only event log that replaces daily memory files. It assumes the current OpenClaw environment already has the Reclaw plugin installed. All memory lives in log.jsonl as structured entries. Extraction happens automatically at session end — you don't write to the log directly. Your job is to state information clearly in conversation so the extraction hook captures it." }, { "title": "How Memory Works", "body": "MEMORY.md is auto-loaded into every session. It has a manual section (goals, preferences) and a generated Reclaw memory snapshot updated nightly.\nReclaw session summary is written into MEMORY.md after each session extraction.\nSubject markdown projections are generated under ~/.openclaw/reclaw/memory/ so OpenClaw can semantically index event-log content through its builtin markdown memory path.\nmemory_search finds entries by keyword, type, subject, or status when the Reclaw plugin has registered that tool in the current OpenClaw environment, and can hit semantic results from MEMORY.md plus generated subject projections.\nmemory_get retrieves a specific entry by ID, reads MEMORY.md, or fetches a full session transcript when the Reclaw plugin has registered that tool in the current OpenClaw environment.\n\nStart with what's already in context (steps 1-3). Only call tools when you need something specific." }, { "title": "Entry Types", "body": "TypeWhat it capturesKey detailtaskAction items, follow-upsHas status: open or donefactUser-specific information learnedPreferences, events, observations, milestonesdecisionA choice with reasoningUse detail for the \"why\"questionAn unresolved open loopResolved by later entries on the same subjectsession_summarySession boundary stateOne per session, summarizes what's in-flight" }, { "title": "Subjects", "body": "Every non-session_summary entry has a subject — a kebab-case slug like auth-migration or reclaw. Subjects are tracked in a registry with a type: project, person, system, or topic (default).\n\nWhen discussing something new, use a clear kebab-case slug. The extraction hook auto-creates subjects it hasn't seen. To explicitly manage subjects:\n\n# List all subjects\nopenclaw reclaw subjects list\n\n# Add a subject with a type\nopenclaw reclaw subjects add auth-migration --type project\nopenclaw reclaw subjects add alice-chen --type person\n\n# Rename a subject (updates registry and all log entries)\nopenclaw reclaw subjects rename old-slug new-slug" }, { "title": "Using memory_search", "body": "Combines structured log filters with keyword search and semantic search over MEMORY.md plus generated subject projections.\n\n# Keyword search\nmemory_search({\"query\": \"webhook retries\"})\n\n# Structured filters\nmemory_search({\"type\": \"decision\", \"subject\": \"auth-migration\"})\nmemory_search({\"type\": \"task\", \"status\": \"open\"})\nmemory_search({\"type\": \"question\"})\n\n# Combined\nmemory_search({\"query\": \"backoff\", \"type\": \"fact\", \"subject\": \"auth-migration\"})\n\nAt least one of query, type, subject, or status is required." }, { "title": "Markdown Projections", "body": "Reclaw keeps one generated markdown file per subject under ~/.openclaw/reclaw/memory/. These files are derived from log.jsonl and exist so OpenClaw's builtin markdown indexer can semantically search event-log content.\n\nTreat projection files as generated output — don't manually edit them\nSuccessful live extraction refreshes touched subject projections automatically\nSuccessful non-dry-run imports refresh the full projection set automatically\nIf the index seems stale, rebuild with openclaw reclaw projection refresh" }, { "title": "Using memory_get", "body": "Three lookup modes based on the path value:\n\n# By entry ID (12-char nanoid from search results)\nmemory_get({\"path\": \"r7Wp3nKx_mZe\"})\n\n# By session transcript (from an entry's session field)\nmemory_get({\"path\": \"session:abc123def456\"})\n\n# By file path\nmemory_get({\"path\": \"MEMORY.md\"})\n\nReading an entry by ID increments its usage score, which helps it persist in the nightly memory snapshot." }, { "title": "Citations", "body": "When referencing a prior event in conversation, cite it as [<12-char-id>] (e.g., [r7Wp3nKx_mZe]). This format is tracked for usage scoring — cited entries are more likely to appear in future memory snapshots." }, { "title": "Corrections and Updates", "body": "The log is append-only. To correct something:\n\nState the correction clearly in conversation. Extraction writes a new entry on the same subject.\nTo mark a task done, say so explicitly. Extraction emits a new task entry with status: \"done\".\nTo answer a question, discuss the resolution. Extraction captures the answer as a fact or decision.\n\nOld entries are never modified. Current state is reconstructed by reading a subject's entries chronologically." }, { "title": "Hard Filter", "body": "Only user-specific information belongs in the log. Ask: \"Would I need to know this person to know this?\" If a general-purpose LLM could produce the content without user context, it should not be extracted. No generic knowledge, no dependency lists, no boilerplate." }, { "title": "CLI Commands", "body": "# Recent log entries\nopenclaw reclaw log\nopenclaw reclaw log --type decision --subject auth-migration --limit 10\n\n# Search with filters\nopenclaw reclaw search \"webhook\"\nopenclaw reclaw search --type task --status open\nopenclaw reclaw search --subject auth-migration --from 2026-02-01 --to 2026-03-01\n\n# Trace a subject's chronological history\nopenclaw reclaw trace\nopenclaw reclaw trace --subject auth-migration\nopenclaw reclaw trace \n\n# Subject management\nopenclaw reclaw subjects list\nopenclaw reclaw subjects add --type \nopenclaw reclaw subjects rename \n\n# Refresh generated subject markdown projections\nopenclaw reclaw projection refresh\nopenclaw reclaw projection list\n\n# Regenerate the MEMORY.md memory snapshot now\nopenclaw reclaw snapshot refresh\n\n# Force-refresh MEMORY.md session summary block from log\nopenclaw reclaw summary refresh\n\n# Import historical conversations\nopenclaw reclaw import \nopenclaw reclaw import status\nopenclaw reclaw import resume \n\n# Setup\nopenclaw reclaw init\nopenclaw reclaw verify\nopenclaw reclaw uninstall" } ], "body": "Reclaw Memory System\n\nReclaw is an append-only event log that replaces daily memory files. It assumes the current OpenClaw environment already has the Reclaw plugin installed. All memory lives in log.jsonl as structured entries. Extraction happens automatically at session end — you don't write to the log directly. Your job is to state information clearly in conversation so the extraction hook captures it.\n\nHow Memory Works\nMEMORY.md is auto-loaded into every session. It has a manual section (goals, preferences) and a generated Reclaw memory snapshot updated nightly.\nReclaw session summary is written into MEMORY.md after each session extraction.\nSubject markdown projections are generated under ~/.openclaw/reclaw/memory/ so OpenClaw can semantically index event-log content through its builtin markdown memory path.\nmemory_search finds entries by keyword, type, subject, or status when the Reclaw plugin has registered that tool in the current OpenClaw environment, and can hit semantic results from MEMORY.md plus generated subject projections.\nmemory_get retrieves a specific entry by ID, reads MEMORY.md, or fetches a full session transcript when the Reclaw plugin has registered that tool in the current OpenClaw environment.\n\nStart with what's already in context (steps 1-3). Only call tools when you need something specific.\n\nEntry Types\nType\tWhat it captures\tKey detail\ntask\tAction items, follow-ups\tHas status: open or done\nfact\tUser-specific information learned\tPreferences, events, observations, milestones\ndecision\tA choice with reasoning\tUse detail for the \"why\"\nquestion\tAn unresolved open loop\tResolved by later entries on the same subject\nsession_summary\tSession boundary state\tOne per session, summarizes what's in-flight\nSubjects\n\nEvery non-session_summary entry has a subject — a kebab-case slug like auth-migration or reclaw. Subjects are tracked in a registry with a type: project, person, system, or topic (default).\n\nWhen discussing something new, use a clear kebab-case slug. The extraction hook auto-creates subjects it hasn't seen. To explicitly manage subjects:\n\n# List all subjects\nopenclaw reclaw subjects list\n\n# Add a subject with a type\nopenclaw reclaw subjects add auth-migration --type project\nopenclaw reclaw subjects add alice-chen --type person\n\n# Rename a subject (updates registry and all log entries)\nopenclaw reclaw subjects rename old-slug new-slug\n\nUsing memory_search\n\nCombines structured log filters with keyword search and semantic search over MEMORY.md plus generated subject projections.\n\n# Keyword search\nmemory_search({\"query\": \"webhook retries\"})\n\n# Structured filters\nmemory_search({\"type\": \"decision\", \"subject\": \"auth-migration\"})\nmemory_search({\"type\": \"task\", \"status\": \"open\"})\nmemory_search({\"type\": \"question\"})\n\n# Combined\nmemory_search({\"query\": \"backoff\", \"type\": \"fact\", \"subject\": \"auth-migration\"})\n\n\nAt least one of query, type, subject, or status is required.\n\nMarkdown Projections\n\nReclaw keeps one generated markdown file per subject under ~/.openclaw/reclaw/memory/. These files are derived from log.jsonl and exist so OpenClaw's builtin markdown indexer can semantically search event-log content.\n\nTreat projection files as generated output — don't manually edit them\nSuccessful live extraction refreshes touched subject projections automatically\nSuccessful non-dry-run imports refresh the full projection set automatically\nIf the index seems stale, rebuild with openclaw reclaw projection refresh\nUsing memory_get\n\nThree lookup modes based on the path value:\n\n# By entry ID (12-char nanoid from search results)\nmemory_get({\"path\": \"r7Wp3nKx_mZe\"})\n\n# By session transcript (from an entry's session field)\nmemory_get({\"path\": \"session:abc123def456\"})\n\n# By file path\nmemory_get({\"path\": \"MEMORY.md\"})\n\n\nReading an entry by ID increments its usage score, which helps it persist in the nightly memory snapshot.\n\nCitations\n\nWhen referencing a prior event in conversation, cite it as [<12-char-id>] (e.g., [r7Wp3nKx_mZe]). This format is tracked for usage scoring — cited entries are more likely to appear in future memory snapshots.\n\nCorrections and Updates\n\nThe log is append-only. To correct something:\n\nState the correction clearly in conversation. Extraction writes a new entry on the same subject.\nTo mark a task done, say so explicitly. Extraction emits a new task entry with status: \"done\".\nTo answer a question, discuss the resolution. Extraction captures the answer as a fact or decision.\n\nOld entries are never modified. Current state is reconstructed by reading a subject's entries chronologically.\n\nHard Filter\n\nOnly user-specific information belongs in the log. Ask: \"Would I need to know this person to know this?\" If a general-purpose LLM could produce the content without user context, it should not be extracted. No generic knowledge, no dependency lists, no boilerplate.\n\nCLI Commands\n# Recent log entries\nopenclaw reclaw log\nopenclaw reclaw log --type decision --subject auth-migration --limit 10\n\n# Search with filters\nopenclaw reclaw search \"webhook\"\nopenclaw reclaw search --type task --status open\nopenclaw reclaw search --subject auth-migration --from 2026-02-01 --to 2026-03-01\n\n# Trace a subject's chronological history\nopenclaw reclaw trace\nopenclaw reclaw trace --subject auth-migration\nopenclaw reclaw trace \n\n# Subject management\nopenclaw reclaw subjects list\nopenclaw reclaw subjects add --type \nopenclaw reclaw subjects rename \n\n# Refresh generated subject markdown projections\nopenclaw reclaw projection refresh\nopenclaw reclaw projection list\n\n# Regenerate the MEMORY.md memory snapshot now\nopenclaw reclaw snapshot refresh\n\n# Force-refresh MEMORY.md session summary block from log\nopenclaw reclaw summary refresh\n\n# Import historical conversations\nopenclaw reclaw import \nopenclaw reclaw import status\nopenclaw reclaw import resume \n\n# Setup\nopenclaw reclaw init\nopenclaw reclaw verify\nopenclaw reclaw uninstall" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/maxpetretta/reclaw", "publisherUrl": "https://clawhub.ai/maxpetretta/reclaw", "owner": "maxpetretta", "version": "2026.3.11-3", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/reclaw", "downloadUrl": "https://openagent3.xyz/downloads/reclaw", "agentUrl": "https://openagent3.xyz/skills/reclaw/agent", "manifestUrl": "https://openagent3.xyz/skills/reclaw/agent.json", "briefUrl": "https://openagent3.xyz/skills/reclaw/agent.md" } }