{
  "schemaVersion": "1.0",
  "item": {
    "slug": "persistent-code-terminal",
    "name": "Persistent Code Terminal",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/Justinzhq/persistent-code-terminal",
    "canonicalUrl": "https://clawhub.ai/Justinzhq/persistent-code-terminal",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/persistent-code-terminal",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=persistent-code-terminal",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      ".github/workflows/ci.yml",
      "CHANGELOG.md",
      "README.md",
      "RELEASE.md",
      "SKILL.md",
      "bin/persistent-code-terminal-attach.sh"
    ],
    "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",
      "slug": "persistent-code-terminal",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-07T01:44:26.333Z",
      "expiresAt": "2026-05-14T01:44:26.333Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=persistent-code-terminal",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=persistent-code-terminal",
        "contentDisposition": "attachment; filename=\"persistent-code-terminal-1.2.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "persistent-code-terminal"
      },
      "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/persistent-code-terminal"
    },
    "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/persistent-code-terminal",
    "agentPageUrl": "https://openagent3.xyz/skills/persistent-code-terminal/agent",
    "manifestUrl": "https://openagent3.xyz/skills/persistent-code-terminal/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/persistent-code-terminal/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": "persistent-code-terminal",
        "body": "A persistent coding terminal for OpenClaw. It creates a tmux session per project:\n\n<project-name>-code-session\n\nThis makes terminal-first AI coding (Codex CLI / Claude Code / any CLI tool) reliable and mobile/SSH friendly:\n\ndetach/reattach without losing output\nkeep shell state across turns\nkeep long-running tasks alive (dev server, watch mode, etc.)"
      },
      {
        "title": "OpenClaw natural-language routing",
        "body": "When user intent is \"use natural language to drive Codex CLI in this repo\", route through this skill first.\n\nIntelligent auto-trigger:\n\nRouting toggle: openclaw.config.dev.autoCodeRouting (default false).\nIf enabled, route coding workflow messages via:\n\n{baseDir}/bin/persistent-code-terminal-route.sh \"<user message>\"\n\n\nThe route script applies:\n\nintent detection (code change / fix / test / build / commit / push / codex phrases)\nsafety filtering (must be git repo + action verbs; bypass on 不要执行，只分析)\nexecution chain:\n\npersistent-code-terminal-auto.sh --max-retries 3 --instruction \"<message>\"\npersistent-code-terminal-summary.sh --lines 120\n\n\nfallback for missing deps (tmux/codex) with doctor guidance.\n\nMulti-project routing:\n\nOne message can include multiple project tasks split by newline / ； / ;\nSupported patterns:\n\n给 <project> 项目 <instruction>\n给<project>项目：<instruction>\n为 <project> 修复/增加/修改 ...\n<project> 项目 ...；<project> 项目 ...\n\n\nExecution is serial per project:\n\nstart.sh --project <project>\nauto.sh --max-retries 3 --instruction \"<instruction>\"\nsummary.sh --lines 120 --json\n\n\nInvalid/unknown project tasks are reported as failed without blocking other tasks.\n\nTrigger shortcuts (recommended):\n\nIf user message starts with codex , treat the remaining text as instruction.\nExecute:\n\n{baseDir}/bin/persistent-code-terminal-codex-exec.sh \"<remaining text>\"\n\n\nThen report with:\n\n{baseDir}/bin/persistent-code-terminal-status.sh\n{baseDir}/bin/persistent-code-terminal-summary.sh --lines 120\n\nPreferred execution pattern in chat-driven runs:\n\nstart.sh (or implicit auto-create from send.sh)\ncodex-exec.sh \"<instruction>\" for Codex-driven work\nstatus.sh / summary.sh for concise progress reporting\n\nIf user says \"检查项目/继续会话/跑测试并总结\", use this skill scripts instead of asking user to type long shell commands."
      },
      {
        "title": "Core model (must follow)",
        "body": "start → send → read → decide\n\nStart/ensure session:\n\n{baseDir}/bin/persistent-code-terminal-start.sh\n\nSend ONE command:\n\n{baseDir}/bin/persistent-code-terminal-send.sh \"<command>\"\n\nAppends a pane sentinel on completion: __PCT_EXIT_CODE__N\nSupports: --timeout <seconds>, --dry-run, --phase <name>\n\nRead output:\n\n{baseDir}/bin/persistent-code-terminal-read.sh\n\nParses the latest sentinel and updates .pct-state.json\n\nCheck state quickly (optional):\n\n{baseDir}/bin/persistent-code-terminal-status.sh\n{baseDir}/bin/persistent-code-terminal-summary.sh --lines 120\n{baseDir}/bin/persistent-code-terminal-doctor.sh\n{baseDir}/bin/persistent-code-terminal-list.sh (list *-code-session)\n{baseDir}/bin/persistent-code-terminal-switch.sh --project <name>\n\nState file:\n\n.pct-state.json (current project directory)\nFields: projectDir, session, lastCommand, lastExitCode, phase, updatedAt\n\nStructured output:\n\n{baseDir}/bin/persistent-code-terminal-read.sh --json\n{baseDir}/bin/persistent-code-terminal-summary.sh --json\n{baseDir}/bin/persistent-code-terminal-auto.sh --json"
      },
      {
        "title": "Codex-first workflow (one-shot)",
        "body": "If Codex CLI is installed as codex, prefer:\n\n{baseDir}/bin/persistent-code-terminal-codex-exec.sh \"<instruction>\"\nDefault behavior uses:\n\ncodex exec --full-auto --sandbox workspace-write --cd <current-dir> \"<instruction>\"\n\n\nYou can pass additional flags before the instruction:\n\n{baseDir}/bin/persistent-code-terminal-codex-exec.sh --json -o /tmp/codex.json \"<instruction>\"\n\n\nSet PCT_CODEX_NO_DEFAULT_FLAGS=1 to disable default flags.\n\nExample:\n\n{baseDir}/bin/persistent-code-terminal-codex-exec.sh \"Implement feature X. Ensure build and tests pass. Commit and push to current branch. Do NOT force push.\""
      },
      {
        "title": "Safety",
        "body": "Never git push --force unless user explicitly requests.\nKeep secrets out of terminal output.\nPrefer feature branches; avoid direct pushes to main/master unless explicitly requested.\nFor network/privileged actions (for example git push), follow active Codex approval/sandbox policy."
      }
    ],
    "body": "persistent-code-terminal\n\nA persistent coding terminal for OpenClaw. It creates a tmux session per project:\n\n<project-name>-code-session\n\nThis makes terminal-first AI coding (Codex CLI / Claude Code / any CLI tool) reliable and mobile/SSH friendly:\n\ndetach/reattach without losing output\nkeep shell state across turns\nkeep long-running tasks alive (dev server, watch mode, etc.)\nOpenClaw natural-language routing\n\nWhen user intent is \"use natural language to drive Codex CLI in this repo\", route through this skill first.\n\nIntelligent auto-trigger:\n\nRouting toggle: openclaw.config.dev.autoCodeRouting (default false).\nIf enabled, route coding workflow messages via:\n{baseDir}/bin/persistent-code-terminal-route.sh \"<user message>\"\nThe route script applies:\nintent detection (code change / fix / test / build / commit / push / codex phrases)\nsafety filtering (must be git repo + action verbs; bypass on 不要执行，只分析)\nexecution chain:\npersistent-code-terminal-auto.sh --max-retries 3 --instruction \"<message>\"\npersistent-code-terminal-summary.sh --lines 120\nfallback for missing deps (tmux/codex) with doctor guidance.\n\nMulti-project routing:\n\nOne message can include multiple project tasks split by newline / ； / ;\nSupported patterns:\n给 <project> 项目 <instruction>\n给<project>项目：<instruction>\n为 <project> 修复/增加/修改 ...\n<project> 项目 ...；<project> 项目 ...\nExecution is serial per project:\nstart.sh --project <project>\nauto.sh --max-retries 3 --instruction \"<instruction>\"\nsummary.sh --lines 120 --json\nInvalid/unknown project tasks are reported as failed without blocking other tasks.\n\nTrigger shortcuts (recommended):\n\nIf user message starts with codex , treat the remaining text as instruction.\nExecute:\n{baseDir}/bin/persistent-code-terminal-codex-exec.sh \"<remaining text>\"\nThen report with:\n{baseDir}/bin/persistent-code-terminal-status.sh\n{baseDir}/bin/persistent-code-terminal-summary.sh --lines 120\n\nPreferred execution pattern in chat-driven runs:\n\nstart.sh (or implicit auto-create from send.sh)\ncodex-exec.sh \"<instruction>\" for Codex-driven work\nstatus.sh / summary.sh for concise progress reporting\n\nIf user says \"检查项目/继续会话/跑测试并总结\", use this skill scripts instead of asking user to type long shell commands.\n\nCore model (must follow)\n\nstart → send → read → decide\n\nStart/ensure session:\n{baseDir}/bin/persistent-code-terminal-start.sh\nSend ONE command:\n{baseDir}/bin/persistent-code-terminal-send.sh \"<command>\"\nAppends a pane sentinel on completion: __PCT_EXIT_CODE__N\nSupports: --timeout <seconds>, --dry-run, --phase <name>\nRead output:\n{baseDir}/bin/persistent-code-terminal-read.sh\nParses the latest sentinel and updates .pct-state.json\nCheck state quickly (optional):\n{baseDir}/bin/persistent-code-terminal-status.sh\n{baseDir}/bin/persistent-code-terminal-summary.sh --lines 120\n{baseDir}/bin/persistent-code-terminal-doctor.sh\n{baseDir}/bin/persistent-code-terminal-list.sh (list *-code-session)\n{baseDir}/bin/persistent-code-terminal-switch.sh --project <name>\n\nState file:\n\n.pct-state.json (current project directory)\nFields: projectDir, session, lastCommand, lastExitCode, phase, updatedAt\n\nStructured output:\n\n{baseDir}/bin/persistent-code-terminal-read.sh --json\n{baseDir}/bin/persistent-code-terminal-summary.sh --json\n{baseDir}/bin/persistent-code-terminal-auto.sh --json\nCodex-first workflow (one-shot)\n\nIf Codex CLI is installed as codex, prefer:\n\n{baseDir}/bin/persistent-code-terminal-codex-exec.sh \"<instruction>\"\nDefault behavior uses:\ncodex exec --full-auto --sandbox workspace-write --cd <current-dir> \"<instruction>\"\nYou can pass additional flags before the instruction:\n{baseDir}/bin/persistent-code-terminal-codex-exec.sh --json -o /tmp/codex.json \"<instruction>\"\nSet PCT_CODEX_NO_DEFAULT_FLAGS=1 to disable default flags.\n\nExample:\n\n{baseDir}/bin/persistent-code-terminal-codex-exec.sh \"Implement feature X. Ensure build and tests pass. Commit and push to current branch. Do NOT force push.\"\nSafety\nNever git push --force unless user explicitly requests.\nKeep secrets out of terminal output.\nPrefer feature branches; avoid direct pushes to main/master unless explicitly requested.\nFor network/privileged actions (for example git push), follow active Codex approval/sandbox policy."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/Justinzhq/persistent-code-terminal",
    "publisherUrl": "https://clawhub.ai/Justinzhq/persistent-code-terminal",
    "owner": "Justinzhq",
    "version": "1.2.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/persistent-code-terminal",
    "downloadUrl": "https://openagent3.xyz/downloads/persistent-code-terminal",
    "agentUrl": "https://openagent3.xyz/skills/persistent-code-terminal/agent",
    "manifestUrl": "https://openagent3.xyz/skills/persistent-code-terminal/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/persistent-code-terminal/agent.md"
  }
}