[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M76YCB53V1BPBQPKSPXWAYPD":3,"skill-same:01M76YCB53V1BPBQPKSPXWAYPD":64},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":41,"commitSha":42,"skillDirectory":43,"forksCount":44,"repositoryHtmlUrl":45,"repositoryHomepageUrl":46,"ownerHtmlUrl":47,"ownerBlogUrl":6,"ownerDescription":48,"updatedAt":49,"install":50,"resources":63},"01M76YCB53V1BPBQPKSPXWAYPD","agentmemory-adapter",null,"Install, wire, audit, and harden AgentMemory for AI coding agents and project workspaces. Use when Codex needs to adopt rohitg00\u002Fagentmemory, configure Codex or OpenClaw memory integration, create ClawHub-ready AgentMemory guidance, debug MCP\u002Fhook\u002Fplugin setup, verify memory capture, or handle edge cases such as npx stalls, sandbox localhost routing, duplicate plugin paths, stale Codex hooks, remote bearer-token security, multi-agent isolation, or missing full-tool proxy mode.","Lord1Egypt\u002Fawesome-skill-forge","Lord1Egypt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F99746359?v=4","01M7592YDNT0XK9WJ0T2MQVPDN","Mohamed Mounir Ali AL Keshita","01M75GRFNHC8VZNWEB8DKDE433",3,"MIT","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fa\u002Fagentmemory-adapter\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: agentmemory-adapter\ndescription: Install, wire, audit, and harden AgentMemory for AI coding agents and project workspaces. Use when Codex needs to adopt rohitg00\u002Fagentmemory, configure Codex or OpenClaw memory integration, create ClawHub-ready AgentMemory guidance, debug MCP\u002Fhook\u002Fplugin setup, verify memory capture, or handle edge cases such as npx stalls, sandbox localhost routing, duplicate plugin paths, stale Codex hooks, remote bearer-token security, multi-agent isolation, or missing full-tool proxy mode.\nlicense: MIT\nmetadata:\n  openclaw:\n    skillKey: agentmemory-adapter\n    homepage: https:\u002F\u002Fgithub.com\u002Fzack-dev-cm\u002Fagentmemory-skills\n    requires:\n      bins:\n        - node\n        - python3\n    envVars:\n      - name: AGENTMEMORY_URL\n        required: false\n        description: Optional AgentMemory REST base URL. Defaults to http:\u002F\u002Flocalhost:3111.\n      - name: AGENTMEMORY_SECRET\n        required: false\n        description: Optional bearer secret for protected AgentMemory servers. Never print this value.\n      - name: AGENTMEMORY_REQUIRE_HTTPS\n        required: false\n        description: Set to 1 to refuse sending bearer tokens over non-loopback HTTP.\n---\n\n# AgentMemory Adapter\n\nUse this skill to make AgentMemory a dependable shared memory layer, not just a copied MCP snippet.\n\n## When To Use\n\nUse this skill when the user wants to install AgentMemory, wire AgentMemory MCP into Codex or OpenClaw, audit an existing memory setup, publish AgentMemory guidance to ClawHub, or debug memory failures involving `npx`, localhost routing, bearer-token safety, stale hooks, duplicate plugin paths, or degraded MCP fallback mode.\n\n## Operating Order\n\n1. Confirm the target surface: Codex, OpenClaw, ClawHub skill packaging, or a specific project workspace.\n2. Read the bundled references for current architecture and platform-specific recipes.\n3. Prefer a pinned local AgentMemory binary over long-lived `npx` commands.\n4. Apply the narrowest config change that satisfies the target surface, preserving existing user settings.\n5. Run the diagnostic script before and after changes, then explain any warnings as operational risks rather than hiding them.\n6. For public skill work, run Codex quick validation, ClawHub skill doctor, leak\u002Fprompt-bleed scans, ClawPatch review, and a gstack-style ship gate before publishing.\n\n## First Pass\n\n1. Inspect the target agent and project scope before editing config. Prefer user-level integration for \"all current projects\"; prefer repo-local integration only when the user asks for project-specific behavior.\n2. Read `references\u002Fagentmemory-deep-analysis.md` for the current AgentMemory architecture and `references\u002Fplatform-recipes.md` for exact Codex, OpenClaw, and ClawHub-ready patterns.\n3. Run `scripts\u002Fcheck_agentmemory.py` before and after changes. Use `--url \u003Cbase-url>` when the server is remote or uses a non-default port.\n4. Preserve existing user config and backups. `agentmemory connect` writes backups under `~\u002F.agentmemory\u002Fbackups`; manual JSON\u002FTOML edits should be equally scoped.\n\n## Install Strategy\n\nPrefer a stable local command over unbounded `npx` in long-lived MCP configs:\n\n```bash\nnpm install --prefix ~\u002F.agentmemory\u002Fnpm @agentmemory\u002Fagentmemory@0.9.22 --omit=optional\nmkdir -p ~\u002F.local\u002Fbin\nln -sfn ~\u002F.agentmemory\u002Fnpm\u002Fnode_modules\u002F.bin\u002Fagentmemory ~\u002F.local\u002Fbin\u002Fagentmemory\n```\n\nUse full optional dependencies only when the user explicitly needs local vector\u002Fimage embeddings and accepts native package install cost. If `npx @agentmemory\u002Fagentmemory@latest` stalls, install the core package locally as above and point MCP entries at `~\u002F.local\u002Fbin\u002Fagentmemory mcp`.\n\n## Codex Workflow\n\n1. Wire MCP with the local command:\n\n```bash\ncodex mcp remove agentmemory 2>\u002Fdev\u002Fnull || true\ncodex mcp add agentmemory \\\n  --env AGENTMEMORY_URL=http:\u002F\u002Flocalhost:3111 \\\n  --env AGENTMEMORY_TOOLS=all \\\n  -- ~\u002F.local\u002Fbin\u002Fagentmemory mcp\n```\n\n2. Install the Codex hook fallback when Codex Desktop or plugin-local hooks are unreliable:\n\n```bash\nagentmemory connect codex --with-hooks --force\n```\n\nThen re-run the MCP add command above if `connect` rewrites the server back to `npx`.\n\n3. Install the upstream Codex plugin or a local marketplace copy when slash skills are wanted. Patch the plugin `.mcp.json` to the local command if avoiding `npx`.\n\n4. Verify:\n\n```bash\ncodex mcp list\npython3 ~\u002F.codex\u002Fskills\u002Fagentmemory-adapter\u002Fscripts\u002Fcheck_agentmemory.py\n```\n\n## OpenClaw Workflow\n\n1. Add the MCP server:\n\n```bash\nopenclaw mcp set agentmemory '{\"command\":\"'\"$HOME\"'\u002F.local\u002Fbin\u002Fagentmemory\",\"args\":[\"mcp\"],\"env\":{\"AGENTMEMORY_URL\":\"http:\u002F\u002Flocalhost:3111\",\"AGENTMEMORY_TOOLS\":\"all\"}}'\n```\n\n2. Install or link the OpenClaw memory plugin from a durable directory, not a temporary source checkout.\n3. Ensure `plugins.slots.memory` is `agentmemory`, `plugins.entries.agentmemory.enabled` is true, and the plugin config includes `base_url`, `token_budget`, `min_confidence`, `fallback_on_error`, and `timeout_ms`.\n4. Verify with:\n\n```bash\nopenclaw mcp show agentmemory\nopenclaw plugins inspect agentmemory\nopenclaw plugins doctor\n```\n\n## Edge-Case Rules\n\n- Treat MCP fallback mode as degraded: it exposes only a small local tool set when the REST server is unreachable. Start the server for the full tool surface.\n- For sandboxed agents, `localhost` may point inside the sandbox. Use a reachable host route and `AGENTMEMORY_FORCE_PROXY=1` only when the route is correct.\n- Never send `AGENTMEMORY_SECRET` over non-loopback `http:\u002F\u002F`. Prefer HTTPS or SSH tunneling; set `AGENTMEMORY_REQUIRE_HTTPS=1` for hard failure.\n- Re-run `agentmemory connect codex --with-hooks` after upgrading AgentMemory because user-scope hooks contain absolute script paths.\n- Remove duplicate OpenClaw plugin paths; stale temp paths can shadow durable installs.\n- Use `AGENT_ID` and `AGENTMEMORY_AGENT_SCOPE=isolated` only when roles need strict separation. Default shared memory is usually better for one user's coding agents.\n- Keep `AGENTMEMORY_INJECT_CONTEXT=false` and `AGENTMEMORY_AUTO_COMPRESS=false` unless the user explicitly accepts token spend.\n- Do not use `memory_governance_delete` or the `\u002Fforget` workflow without explicit confirmation.\n\n## ClawHub-Ready Skill Work\n\nFor a skill intended for ClawHub, keep `SKILL.md` concise, declare OpenClaw runtime requirements in `metadata.openclaw`, avoid paid-skill\u002Flicensing claims, and include a `--clawscan-note` when publishing because this workflow intentionally configures local binaries, MCP, and network access. Use `references\u002Fplatform-recipes.md` before publishing or reviewing the bundle.\n","# AgentMemory Adapter\n\nUse this skill to make AgentMemory a dependable shared memory layer, not just a copied MCP snippet.\n\n## When To Use\n\nUse this skill when the user wants to install AgentMemory, wire AgentMemory MCP into Codex or OpenClaw, audit an existing memory setup, publish AgentMemory guidance to ClawHub, or debug memory failures involving `npx`, localhost routing, bearer-token safety, stale hooks, duplicate plugin paths, or degraded MCP fallback mode.\n\n## Operating Order\n\n1. Confirm the target surface: Codex, OpenClaw, ClawHub skill packaging, or a specific project workspace.\n2. Read the bundled references for current architecture and platform-specific recipes.\n3. Prefer a pinned local AgentMemory binary over long-lived `npx` commands.\n4. Apply the narrowest config change that satisfies the target surface, preserving existing user settings.\n5. Run the diagnostic script before and after changes, then explain any warnings as operational risks rather than hiding them.\n6. For public skill work, run Codex quick validation, ClawHub skill doctor, leak\u002Fprompt-bleed scans, ClawPatch review, and a gstack-style ship gate before publishing.\n\n## First Pass\n\n1. Inspect the target agent and project scope before editing config. Prefer user-level integration for \"all current projects\"; prefer repo-local integration only when the user asks for project-specific behavior.\n2. Read `references\u002Fagentmemory-deep-analysis.md` for the current AgentMemory architecture and `references\u002Fplatform-recipes.md` for exact Codex, OpenClaw, and ClawHub-ready patterns.\n3. Run `scripts\u002Fcheck_agentmemory.py` before and after changes. Use `--url \u003Cbase-url>` when the server is remote or uses a non-default port.\n4. Preserve existing user config and backups. `agentmemory connect` writes backups under `~\u002F.agentmemory\u002Fbackups`; manual JSON\u002FTOML edits should be equally scoped.\n\n## Install Strategy\n\nPrefer a stable local command over unbounded `npx` in long-lived MCP configs:\n\n```bash\nnpm install --prefix ~\u002F.agentmemory\u002Fnpm @agentmemory\u002Fagentmemory@0.9.22 --omit=optional\nmkdir -p ~\u002F.local\u002Fbin\nln -sfn ~\u002F.agentmemory\u002Fnpm\u002Fnode_modules\u002F.bin\u002Fagentmemory ~\u002F.local\u002Fbin\u002Fagentmemory\n```\n\nUse full optional dependencies only when the user explicitly needs local vector\u002Fimage embeddings and accepts native package install cost. If `npx @agentmemory\u002Fagentmemory@latest` stalls, install the core package locally as above and point MCP entries at `~\u002F.local\u002Fbin\u002Fagentmemory mcp`.\n\n## Codex Workflow\n\n1. Wire MCP with the local command:\n\n```bash\ncodex mcp remove agentmemory 2>\u002Fdev\u002Fnull || true\ncodex mcp add agentmemory \\\n  --env AGENTMEMORY_URL=http:\u002F\u002Flocalhost:3111 \\\n  --env AGENTMEMORY_TOOLS=all \\\n  -- ~\u002F.local\u002Fbin\u002Fagentmemory mcp\n```\n\n2. Install the Codex hook fallback when Codex Desktop or plugin-local hooks are unreliable:\n\n```bash\nagentmemory connect codex --with-hooks --force\n```\n\nThen re-run the MCP add command above if `connect` rewrites the server back to `npx`.\n\n3. Install the upstream Codex plugin or a local marketplace copy when slash skills are wanted. Patch the plugin `.mcp.json` to the local command if avoiding `npx`.\n\n4. Verify:\n\n```bash\ncodex mcp list\npython3 ~\u002F.codex\u002Fskills\u002Fagentmemory-adapter\u002Fscripts\u002Fcheck_agentmemory.py\n```\n\n## OpenClaw Workflow\n\n1. Add the MCP server:\n\n```bash\nopenclaw mcp set agentmemory '{\"command\":\"'\"$HOME\"'\u002F.local\u002Fbin\u002Fagentmemory\",\"args\":[\"mcp\"],\"env\":{\"AGENTMEMORY_URL\":\"http:\u002F\u002Flocalhost:3111\",\"AGENTMEMORY_TOOLS\":\"all\"}}'\n```\n\n2. Install or link the OpenClaw memory plugin from a durable directory, not a temporary source checkout.\n3. Ensure `plugins.slots.memory` is `agentmemory`, `plugins.entries.agentmemory.enabled` is true, and the plugin config includes `base_url`, `token_budget`, `min_confidence`, `fallback_on_error`, and `timeout_ms`.\n4. Verify with:\n\n```bash\nopenclaw mcp show agentmemory\nopenclaw plugins inspect agentmemory\nopenclaw plugins doctor\n```\n\n## Edge-Case Rules\n\n- Treat MCP fallback mode as degraded: it exposes only a small local tool set when the REST server is unreachable. Start the server for the full tool surface.\n- For sandboxed agents, `localhost` may point inside the sandbox. Use a reachable host route and `AGENTMEMORY_FORCE_PROXY=1` only when the route is correct.\n- Never send `AGENTMEMORY_SECRET` over non-loopback `http:\u002F\u002F`. Prefer HTTPS or SSH tunneling; set `AGENTMEMORY_REQUIRE_HTTPS=1` for hard failure.\n- Re-run `agentmemory connect codex --with-hooks` after upgrading AgentMemory because user-scope hooks contain absolute script paths.\n- Remove duplicate OpenClaw plugin paths; stale temp paths can shadow durable installs.\n- Use `AGENT_ID` and `AGENTMEMORY_AGENT_SCOPE=isolated` only when roles need strict separation. Default shared memory is usually better for one user's coding agents.\n- Keep `AGENTMEMORY_INJECT_CONTEXT=false` and `AGENTMEMORY_AUTO_COMPRESS=false` unless the user explicitly accepts token spend.\n- Do not use `memory_governance_delete` or the `\u002Fforget` workflow without explicit confirmation.\n\n## ClawHub-Ready Skill Work\n\nFor a skill intended for ClawHub, keep `SKILL.md` concise, declare OpenClaw runtime requirements in `metadata.openclaw`, avoid paid-skill\u002Flicensing claims, and include a `--clawscan-note` when publishing because this workflow intentionally configures local binaries, MCP, and network access. Use `references\u002Fplatform-recipes.md` before publishing or reviewing the bundle.\n",{"name":5,"license":15,"metadata":24,"description":7},{"openclaw":25},{"envVars":26,"homepage":36,"requires":37,"skillKey":5},[27,30,33],{"name":28,"required":20,"description":29},"AGENTMEMORY_URL","Optional AgentMemory REST base URL. Defaults to http:\u002F\u002Flocalhost:3111.",{"name":31,"required":20,"description":32},"AGENTMEMORY_SECRET","Optional bearer secret for protected AgentMemory servers. Never print this value.",{"name":34,"required":20,"description":35},"AGENTMEMORY_REQUIRE_HTTPS","Set to 1 to refuse sending bearer tokens over non-loopback HTTP.","https:\u002F\u002Fgithub.com\u002Fzack-dev-cm\u002Fagentmemory-skills",{"bins":38},[39,40],"node","python3","master","4aa004da461d820f4ea5499e58eb8d196cf99016","community\u002Fclawhub\u002Fa\u002Fagentmemory-adapter",0,"https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge","https:\u002F\u002Fskill-explorer-rho.vercel.app","https:\u002F\u002Fgithub.com\u002FLord1Egypt","Unbreakable. Unstoppable. Unmatched⚡.\r\n Speaker of machine tongues 🤖  | World-builder of decentralized dreams 🌐 | Defender of the digital realm🛡️.","2026-06-10 08:21:01",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agentmemory-adapter",{"tool":56,"command":57},"bunx","bunx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agentmemory-adapter",{"tool":59,"command":60},"pnpm","pnpm dlx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agentmemory-adapter","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fa\u002Fagentmemory-adapter\u002FSKILL.md",true,[],[65,73,80,87,94,101],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":72},"01M785WKJB99G3STR1NF9NRTJW","youtube-content","YouTube transcripts to summaries, threads, blogs.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fyoutube-content\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":72},"01M785WEFHB0G2S9YA6EPR3H0E","xurl","X\u002FTwitter via xurl CLI: post, search, DM, media, v2 API.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fxurl\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":72},"01M785W98EDN4EFJXJPMQEHCRC","touchdesigner-mcp","Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftouchdesigner-mcp\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":72},"01M785W5S9FP3QBH2V191E025P","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":72},"01M785W2AS08NSSH9YATCV6FJB","teams-meeting-pipeline","Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fteams-meeting-pipeline\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":72},"01M785VXKW4T34BTP1SCHW1R7F","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fsystematic-debugging\u002FSKILL.md",[],[]]