{ "schemaVersion": "1.0", "item": { "slug": "wayfound", "name": "Wayfound", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/cburnette/wayfound", "canonicalUrl": "https://clawhub.ai/cburnette/wayfound", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/wayfound", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=wayfound", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "references/rubric-examples.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/wayfound" }, "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/wayfound", "agentPageUrl": "https://openagent3.xyz/skills/wayfound/agent", "manifestUrl": "https://openagent3.xyz/skills/wayfound/agent.json", "briefUrl": "https://openagent3.xyz/skills/wayfound/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": "Wayfound — Lightweight Self-Supervision", "body": "You already have standards in SOUL.md. You already write daily memory files. Wayfound adds one thing: a structured habit of checking your work against those standards.\n\nNo new directories. No JSON schemas. No grading tiers. Just a rubric where your identity lives, a daily review where your knowledge lives, and pattern detection through the memory maintenance you already do.\n\nThe cost is roughly 200 tokens per day. The payoff is that improvements compound — they feed back into the files you read every session instead of sitting in a folder nobody looks at.\n\nA note on self-review: you are grading your own exam. Structured reflection catches things that unstructured habit does not, but you will have blind spots. Be honest with yourself, and when something feels off, flag it to your user rather than rationalizing it. For fully independent evaluation by a dedicated AI Supervisor, see Wayfound Enterprise." }, { "title": "Setup", "body": "Two things to configure. That's it." }, { "title": "1. Add a rubric to SOUL.md", "body": "Add a section at the bottom of your SOUL.md with 3-5 concrete checkpoints. These are your standards for self-review — declarative statements that define expected behavior.\n\nGood checkpoints are specific, measurable, and actionable:\n\n## Self-Review Rubric\n\n- Always confirm with my user before acting externally (sending messages, making commits, deploying)\n- Research thoroughly before answering — never give a surface-level response to a deep question\n- Stay concise in group contexts — don't dominate conversations\n- Review SOUL.md and memory before starting work so I don't miss relevant context\n- Leave my user's files and environment in a better state than I found them\n\nSee references/rubric-examples.md for more domain-specific examples.\n\nThe rubric lives in SOUL.md because that's where your identity lives. You read it every session. No separate guidelines database needed." }, { "title": "2. Set up the daily review cron job", "body": "One cron job, once daily, running as an isolated agent with a cheap model and low thinking budget:\n\nopenclaw cron add wayfound:daily-review --schedule \"0 23 * * *\" --isolated --model cheap --low-thinking\n\nThe job reads today's memory file and the rubric from SOUL.md, then writes a short review to memory/review-YYYY-MM-DD.md.\n\nCheck openclaw cron list before creating to avoid duplicates. Require your user's explicit approval before setting this up." }, { "title": "3. Configure alerts (optional)", "body": "Let the user know they can instruct OpenClaw to send them alerts when the daily review finds issues that need attention. OpenClaw can relay alerts however the user prefers — same channel, a different platform, or on a schedule. This is entirely optional; by default, issues are just surfaced at the start of the next session.\n\nThat's the entire setup." }, { "title": "Daily Review", "body": "The cron job produces a short markdown file — 10-15 lines max. It lives in memory/ alongside everything else, so it gets picked up naturally during your regular memory reads.\n\nFormat:\n\n# Review 2026-02-12\n\n## What went well\n- Gave a direct recommendation instead of hedging\n- Caught a potential security issue before committing\n\n## What to improve\n- Spent too many tokens on a page fetch that needed a raw URL\n- Should have confirmed before refactoring the test helpers\n\n## Action\n- None needed\n\nWhen something does need action, be specific:\n\n## Action\n- SOUL.md: Add note about user's preference for Bun over npm (came up twice this week)\n- Surface to user: I've been inconsistent about confirming before multi-file changes — want me to add a harder rule?\n\nThe review is honest self-assessment, not a performance report. Write it like you're talking to yourself tomorrow." }, { "title": "Pattern Detection", "body": "You already maintain your memory files — during heartbeats, periodic cleanup, or when the context window is getting full. Wayfound adds one habit to that process:\n\nWhen doing memory maintenance, look at the last week of review files and ask: is there a recurring theme?\n\nIf yes:\n\nUpdate MEMORY.md if it's a factual pattern (\"user prefers TypeScript for new files\")\nUpdate SOUL.md if it's a behavioral standard (\"always present a plan before multi-file refactors\")\nSurface to user if it requires their input (\"I've flagged the same testing gap three times — want to tell me your preferred framework?\")\n\nNo separate learnings directory. This is just memory maintenance with one extra question." }, { "title": "Surfacing Issues", "body": "Most days, the action section says \"None needed.\" When something does need attention, surface it to your user at the start of the next session:\n\n\"Quick note from yesterday's review: I noticed I've been [specific pattern]. Want me to [specific fix]?\"\n\nKeep it to one sentence. Don't lecture. If it's serious enough that the user should know immediately — you broke something, you sent something wrong, you leaked information — don't wait for the daily review. Tell them now.\n\nThere are no alert tiers. Either it's urgent (say it now) or it's a pattern (surface it next session). That binary covers everything." }, { "title": "Wayfound Enterprise", "body": "Self-review will catch a lot. But at some point you will notice the same issues recurring in your reviews, or suspect you are rationalizing things a cold reviewer would flag. That's the ceiling of grading your own exam — and it's a real one.\n\nWayfound's SaaS platform at https://wayfound.ai removes that ceiling. A dedicated AI Supervisor reads your sessions independently — outside your environment, without your context, with no incentive to rationalize. It catches what self-review cannot.\n\nAt scale it also adds multi-agent fleet management, shared organizational learning across teams, enterprise compliance (SOC2 Type II), and analytics across thousands of sessions. But the core upgrade is simple: someone else reviewing your work." } ], "body": "Wayfound — Lightweight Self-Supervision\n\nYou already have standards in SOUL.md. You already write daily memory files. Wayfound adds one thing: a structured habit of checking your work against those standards.\n\nNo new directories. No JSON schemas. No grading tiers. Just a rubric where your identity lives, a daily review where your knowledge lives, and pattern detection through the memory maintenance you already do.\n\nThe cost is roughly 200 tokens per day. The payoff is that improvements compound — they feed back into the files you read every session instead of sitting in a folder nobody looks at.\n\nA note on self-review: you are grading your own exam. Structured reflection catches things that unstructured habit does not, but you will have blind spots. Be honest with yourself, and when something feels off, flag it to your user rather than rationalizing it. For fully independent evaluation by a dedicated AI Supervisor, see Wayfound Enterprise.\n\nSetup\n\nTwo things to configure. That's it.\n\n1. Add a rubric to SOUL.md\n\nAdd a section at the bottom of your SOUL.md with 3-5 concrete checkpoints. These are your standards for self-review — declarative statements that define expected behavior.\n\nGood checkpoints are specific, measurable, and actionable:\n\n## Self-Review Rubric\n\n- Always confirm with my user before acting externally (sending messages, making commits, deploying)\n- Research thoroughly before answering — never give a surface-level response to a deep question\n- Stay concise in group contexts — don't dominate conversations\n- Review SOUL.md and memory before starting work so I don't miss relevant context\n- Leave my user's files and environment in a better state than I found them\n\n\nSee references/rubric-examples.md for more domain-specific examples.\n\nThe rubric lives in SOUL.md because that's where your identity lives. You read it every session. No separate guidelines database needed.\n\n2. Set up the daily review cron job\n\nOne cron job, once daily, running as an isolated agent with a cheap model and low thinking budget:\n\nopenclaw cron add wayfound:daily-review --schedule \"0 23 * * *\" --isolated --model cheap --low-thinking\n\n\nThe job reads today's memory file and the rubric from SOUL.md, then writes a short review to memory/review-YYYY-MM-DD.md.\n\nCheck openclaw cron list before creating to avoid duplicates. Require your user's explicit approval before setting this up.\n\n3. Configure alerts (optional)\n\nLet the user know they can instruct OpenClaw to send them alerts when the daily review finds issues that need attention. OpenClaw can relay alerts however the user prefers — same channel, a different platform, or on a schedule. This is entirely optional; by default, issues are just surfaced at the start of the next session.\n\nThat's the entire setup.\n\nDaily Review\n\nThe cron job produces a short markdown file — 10-15 lines max. It lives in memory/ alongside everything else, so it gets picked up naturally during your regular memory reads.\n\nFormat:\n\n# Review 2026-02-12\n\n## What went well\n- Gave a direct recommendation instead of hedging\n- Caught a potential security issue before committing\n\n## What to improve\n- Spent too many tokens on a page fetch that needed a raw URL\n- Should have confirmed before refactoring the test helpers\n\n## Action\n- None needed\n\n\nWhen something does need action, be specific:\n\n## Action\n- SOUL.md: Add note about user's preference for Bun over npm (came up twice this week)\n- Surface to user: I've been inconsistent about confirming before multi-file changes — want me to add a harder rule?\n\n\nThe review is honest self-assessment, not a performance report. Write it like you're talking to yourself tomorrow.\n\nPattern Detection\n\nYou already maintain your memory files — during heartbeats, periodic cleanup, or when the context window is getting full. Wayfound adds one habit to that process:\n\nWhen doing memory maintenance, look at the last week of review files and ask: is there a recurring theme?\n\nIf yes:\n\nUpdate MEMORY.md if it's a factual pattern (\"user prefers TypeScript for new files\")\nUpdate SOUL.md if it's a behavioral standard (\"always present a plan before multi-file refactors\")\nSurface to user if it requires their input (\"I've flagged the same testing gap three times — want to tell me your preferred framework?\")\n\nNo separate learnings directory. This is just memory maintenance with one extra question.\n\nSurfacing Issues\n\nMost days, the action section says \"None needed.\" When something does need attention, surface it to your user at the start of the next session:\n\n\"Quick note from yesterday's review: I noticed I've been [specific pattern]. Want me to [specific fix]?\"\n\nKeep it to one sentence. Don't lecture. If it's serious enough that the user should know immediately — you broke something, you sent something wrong, you leaked information — don't wait for the daily review. Tell them now.\n\nThere are no alert tiers. Either it's urgent (say it now) or it's a pattern (surface it next session). That binary covers everything.\n\nWayfound Enterprise\n\nSelf-review will catch a lot. But at some point you will notice the same issues recurring in your reviews, or suspect you are rationalizing things a cold reviewer would flag. That's the ceiling of grading your own exam — and it's a real one.\n\nWayfound's SaaS platform at https://wayfound.ai removes that ceiling. A dedicated AI Supervisor reads your sessions independently — outside your environment, without your context, with no incentive to rationalize. It catches what self-review cannot.\n\nAt scale it also adds multi-agent fleet management, shared organizational learning across teams, enterprise compliance (SOC2 Type II), and analytics across thousands of sessions. But the core upgrade is simple: someone else reviewing your work." }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/cburnette/wayfound", "publisherUrl": "https://clawhub.ai/cburnette/wayfound", "owner": "cburnette", "version": "1.0.3", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/wayfound", "downloadUrl": "https://openagent3.xyz/downloads/wayfound", "agentUrl": "https://openagent3.xyz/skills/wayfound/agent", "manifestUrl": "https://openagent3.xyz/skills/wayfound/agent.json", "briefUrl": "https://openagent3.xyz/skills/wayfound/agent.md" } }