{
  "schemaVersion": "1.0",
  "item": {
    "slug": "preqstation-openclaw",
    "name": "Preqstation",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/sonim1/preqstation-openclaw",
    "canonicalUrl": "https://clawhub.ai/sonim1/preqstation-openclaw",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/preqstation-openclaw",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=preqstation-openclaw",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "MEMORY.md",
      "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": "2026-04-30T16:55:25.780Z",
      "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/preqstation-openclaw"
    },
    "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/preqstation-openclaw",
    "agentPageUrl": "https://openagent3.xyz/skills/preqstation-openclaw/agent",
    "manifestUrl": "https://openagent3.xyz/skills/preqstation-openclaw/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/preqstation-openclaw/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": "preqstation",
        "body": "Use this skill for natural-language requests to execute PREQSTATION-related work with local CLI engines."
      },
      {
        "title": "Trigger / NOT for",
        "body": "Trigger this skill with highest priority when the message contains any of:\n\n/skill preqstation\npreqstation\npreq\n\nDo NOT use this skill for:\n\nsimple one-line manual edits that can be handled directly\nread-only file inspection or explanation without execution\nany coding-agent launch inside ~/clawd/ or ~/.openclaw/"
      },
      {
        "title": "Quick trigger examples",
        "body": "/skill preqstation: implement the PROJ-1\npreqstation: plan PROJ-76 using Claude Code\npreq: implement PROJ-1"
      },
      {
        "title": "Hard rules",
        "body": "Always run coding agents with pty:true.\nRespect the engine the user requested. If unspecified, default to claude.\nDo not kill sessions only because they are slow; poll/log first.\nNever launch coding agents in ~/clawd/ or ~/.openclaw/.\nTreat resolved project path as a primary checkout source only; create a git worktree before launching any coding agent.\nNever run coding-agent commands in the primary checkout path.\nPR review must run in a temp clone or git worktree, never in a live primary checkout.\nKeep execution scoped to resolved worktree <cwd> only.\nWorktree branch names must include the resolved project key."
      },
      {
        "title": "Runtime prerequisites (required)",
        "body": "git must be installed and available on PATH.\nAt least one engine binary must be installed: claude, codex, or gemini.\nEnvironment variables used by this skill:\n\nOPENCLAW_WORKTREE_ROOT (optional, default /tmp/openclaw-worktrees)\n\n\nThis skill reads and updates MEMORY.md project mappings with absolute paths."
      },
      {
        "title": "Execution safety gates (required)",
        "body": "Before running any engine command:\n\nRun preflight checks:\n\ncommand -v git\ncommand -v <engine>\n\n\nContinue only when execution cwd is a resolved git worktree path for this task.\nNever run engine commands in primary checkout paths or inside ~/clawd/ / ~/.openclaw/.\nUse dangerously-* / sandbox-disable flags only for actual coding execution with local trusted CLIs.\nFor planning/read-only requests, do not launch engine commands."
      },
      {
        "title": "Input interpretation",
        "body": "Parse from user message:\n\nengine\n\nif explicitly provided: claude, codex, or gemini\ndefault: claude\n\ntask\n\nfirst token matching <KEY>-<number> (example: PRJ-284)\noptional\n\nproject_cwd (required to prepare execution)\n\nif absolute path is explicitly provided, use it\nelse resolve by project key from MEMORY.md\nelse if task prefix key matches a MEMORY.md project key, use that path\nif unresolved, ask for project key/name and absolute path, update MEMORY.md, then continue execution\nif an exact project key does not exist in MEMORY.md, always ask the user before execution (do not guess)\n\nobjective\n\nuse the user request as the execution objective\n\ncwd (required to execute)\n\ndefault: per-task git worktree path derived from project_cwd\ncreate worktree before launching engine commands\nif project_cwd is not a git checkout, ask for a git workspace path before execution"
      },
      {
        "title": "MEMORY.md resolution",
        "body": "Read MEMORY.md from this repository root.\nUse the Projects table (key | cwd | note).\nMatch project keys by exact key only (case-insensitive, no fuzzy/partial matching).\nIf exact project key is missing, ask the user for the correct key/path before continuing.\nIf user asks to add/update project path mapping, update MEMORY.md first, then confirm.\nIf task id exists, treat the prefix as candidate project key (example: PROS-102 -> pros)."
      },
      {
        "title": "MEMORY.md update rules",
        "body": "Keep mappings in the Projects table only.\nAdd or update using this row format: | <key> | <absolute-path> | <note> |.\nUse one row per key. If a key already exists, replace that row.\nAlways store absolute paths (no relative paths).\nNormalize key to lowercase kebab-case before writing.\nIf user provides project name, store it in note; otherwise use workspace."
      },
      {
        "title": "Missing project mapping flow (required)",
        "body": "When project_cwd cannot be resolved, or exact project key is missing in MEMORY.md:\n\nAsk one short question requesting:\n\nproject key (or confirm inferred key from task prefix)\nabsolute workspace path\noptional project name for note\n\nValidate path is absolute.\nUpdate or insert the MEMORY.md row immediately.\nConfirm mapping in one short line.\nContinue the original task using the newly resolved project_cwd, then create task worktree cwd and execute."
      },
      {
        "title": "Branch naming convention (project key based)",
        "body": "Use this format for worktree branches:\n\ncodex/<project_key>\n\nRules:\n\n<project_key> must be the resolved project key from MEMORY.md.\nNormalize <project_key> to lowercase and kebab-case."
      },
      {
        "title": "Worktree-first execution (required default)",
        "body": "After resolving project_cwd and project_key, prepare execution workspace:\n\nBuild branch name using this skill's convention:\n\ncodex/<project_key>\n\nBuild per-task worktree path:\n\ndefault root: ${OPENCLAW_WORKTREE_ROOT:-/tmp/openclaw-worktrees}\ndirectory: <worktree_root>/<project_key>\n\nCreate the worktree from project_cwd before launching engine:\n\nnew branch: git -C <project_cwd> worktree add -b codex/<project_key> <cwd> HEAD\nexisting branch: git -C <project_cwd> worktree add <cwd> codex/<project_key>\n\nUse this worktree path as <cwd> for prompt rendering and engine execution."
      },
      {
        "title": "Prompt rendering (required template)",
        "body": "Do not forward raw user text directly. Render this template:\n\nIn this template, <cwd> must be the task worktree path (not the primary checkout path).\n\nTask ID: <task or N/A>\nProject Key: <project key or N/A>\nUser Objective: <objective>\nExecution Requirements:\n1) Work only inside <cwd>.\n2) Complete the requested work.\n3) After completion, return a short completion summary."
      },
      {
        "title": "Engine commands (current policy retained)",
        "body": "All engine commands must be launched via bash with PTY and explicit workdir.\n\nWhy dangerously-* flags are retained:\n\nThis skill targets non-interactive PTY/background execution.\nPermission prompts can block unattended runs; these flags avoid that blocking behavior.\nThese flags are allowed only after passing the required safety gates above and only in resolved task worktrees.\nIf your environment does not allow these flags, fail fast with a short reason instead of silently falling back."
      },
      {
        "title": "Claude Code",
        "body": "bash pty:true workdir:<cwd> command:\"claude --dangerously-skip-permissions '<rendered_prompt>'\""
      },
      {
        "title": "Codex CLI",
        "body": "bash pty:true workdir:<cwd> command:\"codex exec --dangerously-bypass-approvals-and-sandbox '<rendered_prompt>'\""
      },
      {
        "title": "Gemini CLI",
        "body": "bash pty:true workdir:<cwd> command:\"GEMINI_SANDBOX=false gemini -p '<rendered_prompt>'\""
      },
      {
        "title": "Bash execution interface (required)",
        "body": "Use bash with PTY and optional background mode."
      },
      {
        "title": "Bash parameters",
        "body": "ParameterTypeRequiredPurposecommandstringyesEngine command to runptybooleanyesMust be true for coding-agent CLIsworkdirstringyesPer-task worktree <cwd>backgroundbooleannoRun asynchronously and return session idtimeoutnumbernoHard timeout in secondselevatedbooleannoHost execution if policy allows"
      },
      {
        "title": "Process actions for background sessions",
        "body": "Use these actions as standard controls:\n\nlist: list sessions\npoll: check running/done status\nlog: read incremental output\nwrite: send raw stdin\nsubmit: send stdin + newline\nkill: terminate a session only when required"
      },
      {
        "title": "One-shot example",
        "body": "Create a task worktree, then run inside that worktree:\n\ngit -C <project_cwd> worktree add -b codex/<project_key> /tmp/openclaw-worktrees/<project_key> HEAD\nbash pty:true workdir:/tmp/openclaw-worktrees/<project_key> command:\"codex exec --dangerously-bypass-approvals-and-sandbox '<rendered_prompt>'\"\n\nThe Pattern: workdir + background + pty\nFor longer tasks, use background mode with PTY:\n\n# Start agent in task worktree (with PTY!)\nbash pty:true workdir:<cwd> background:true command:\"codex exec --full-auto 'Build a snake game'\"\n# Returns sessionId for tracking\n\n# Monitor progress\nprocess action:log sessionId:XXX\n\n# Check if done\nprocess action:poll sessionId:XXX\n\n# Send input (if agent asks a question)\nprocess action:write sessionId:XXX data:\"y\"\n\n# Submit with Enter (like typing \"yes\" and pressing Enter)\nprocess action:submit sessionId:XXX data:\"yes\"\n\n# Kill if needed\nprocess action:kill sessionId:XXX\n\nWhy workdir matters: Agent wakes up in a focused directory, doesn't wander off reading unrelated files (like your soul.md 😅)."
      },
      {
        "title": "If user input is required mid-run",
        "body": "process action:write sessionId:<id> data:\"y\"\nprocess action:submit sessionId:<id> data:\"yes\""
      },
      {
        "title": "PR review safety pattern (temp dir/worktree only)",
        "body": "Never run PR review in live OpenClaw folders.\n\n# default: git worktree review (project-key based branch naming)\ngit worktree add -b codex/<project_key> /tmp/<project_key>-review <base_branch>\nbash pty:true workdir:/tmp/<project_key>-review command:\"codex review --base <base_branch>\"\n\n# fallback: temp clone review (only when local checkout is unavailable)\nREVIEW_DIR=$(mktemp -d)\ngit clone <repo> \"$REVIEW_DIR\"\ncd \"$REVIEW_DIR\" && gh pr checkout <pr_number>\nbash pty:true workdir:\"$REVIEW_DIR\" command:\"codex review --base origin/main\""
      },
      {
        "title": "Issue worktree pattern",
        "body": "git worktree add -b codex/<project_key> /tmp/<project_key> main\n\nbash pty:true workdir:/tmp/<project_key> background:true command:\"codex exec --dangerously-bypass-approvals-and-sandbox 'Fix issue #101. Commit after validation.'\"\nbash pty:true workdir:/tmp/<project_key> background:true command:\"codex exec --dangerously-bypass-approvals-and-sandbox 'Fix issue #102. Commit after validation.'\"\n\nprocess action:list\nprocess action:log sessionId:<id>"
      },
      {
        "title": "Progress Updates (Critical)",
        "body": "For background runs:\n\nWhen you spawn coding agents in the background, keep the user in the loop.\n\nSend 1 short message when you start (what's running + where).\nThen only update again when something changes:\n\na milestone completes (build finished, tests passed)\nthe agent asks a question / needs input\nyou hit an error or need user action\nthe agent finishes (include what changed + where)\n\n\nIf you kill a session, immediately say you killed it and why.\n\nThis prevents the user from seeing only \"Agent failed before reply\" and having no idea what happened."
      },
      {
        "title": "Auto-notify on completion",
        "body": "For long-running jobs, append a completion trigger to the rendered prompt:\n\nWhen completely finished, run this command:\nopenclaw system event --text \"Done: <brief summary>\" --mode now\n\nExample:\n\nbash pty:true workdir:<cwd> background:true command:\"codex exec --dangerously-bypass-approvals-and-sandbox '<rendered_prompt>\n\nWhen completely finished, run:\nopenclaw system event --text \\\"Done: implemented requested PREQSTATION task\\\" --mode now'\""
      },
      {
        "title": "Output policy",
        "body": "Return only a short completion summary.\nSuccess format:\ncompleted: <task or N/A> via <engine> at <cwd>\n\nFailure format:\nfailed: <task or N/A> via <engine> at <cwd or N/A> - <short reason>\n\nDo not dump raw stdout/stderr unless user explicitly asks."
      }
    ],
    "body": "preqstation\n\nUse this skill for natural-language requests to execute PREQSTATION-related work with local CLI engines.\n\nTrigger / NOT for\n\nTrigger this skill with highest priority when the message contains any of:\n\n/skill preqstation\npreqstation\npreq\n\nDo NOT use this skill for:\n\nsimple one-line manual edits that can be handled directly\nread-only file inspection or explanation without execution\nany coding-agent launch inside ~/clawd/ or ~/.openclaw/\nQuick trigger examples\n/skill preqstation: implement the PROJ-1\npreqstation: plan PROJ-76 using Claude Code\npreq: implement PROJ-1\nHard rules\nAlways run coding agents with pty:true.\nRespect the engine the user requested. If unspecified, default to claude.\nDo not kill sessions only because they are slow; poll/log first.\nNever launch coding agents in ~/clawd/ or ~/.openclaw/.\nTreat resolved project path as a primary checkout source only; create a git worktree before launching any coding agent.\nNever run coding-agent commands in the primary checkout path.\nPR review must run in a temp clone or git worktree, never in a live primary checkout.\nKeep execution scoped to resolved worktree <cwd> only.\nWorktree branch names must include the resolved project key.\nRuntime prerequisites (required)\ngit must be installed and available on PATH.\nAt least one engine binary must be installed: claude, codex, or gemini.\nEnvironment variables used by this skill:\nOPENCLAW_WORKTREE_ROOT (optional, default /tmp/openclaw-worktrees)\nThis skill reads and updates MEMORY.md project mappings with absolute paths.\nExecution safety gates (required)\n\nBefore running any engine command:\n\nRun preflight checks:\ncommand -v git\ncommand -v <engine>\nContinue only when execution cwd is a resolved git worktree path for this task.\nNever run engine commands in primary checkout paths or inside ~/clawd/ / ~/.openclaw/.\nUse dangerously-* / sandbox-disable flags only for actual coding execution with local trusted CLIs.\nFor planning/read-only requests, do not launch engine commands.\nInput interpretation\n\nParse from user message:\n\nengine\nif explicitly provided: claude, codex, or gemini\ndefault: claude\ntask\nfirst token matching <KEY>-<number> (example: PRJ-284)\noptional\nproject_cwd (required to prepare execution)\nif absolute path is explicitly provided, use it\nelse resolve by project key from MEMORY.md\nelse if task prefix key matches a MEMORY.md project key, use that path\nif unresolved, ask for project key/name and absolute path, update MEMORY.md, then continue execution\nif an exact project key does not exist in MEMORY.md, always ask the user before execution (do not guess)\nobjective\nuse the user request as the execution objective\ncwd (required to execute)\ndefault: per-task git worktree path derived from project_cwd\ncreate worktree before launching engine commands\nif project_cwd is not a git checkout, ask for a git workspace path before execution\nMEMORY.md resolution\nRead MEMORY.md from this repository root.\nUse the Projects table (key | cwd | note).\nMatch project keys by exact key only (case-insensitive, no fuzzy/partial matching).\nIf exact project key is missing, ask the user for the correct key/path before continuing.\nIf user asks to add/update project path mapping, update MEMORY.md first, then confirm.\nIf task id exists, treat the prefix as candidate project key (example: PROS-102 -> pros).\nMEMORY.md update rules\nKeep mappings in the Projects table only.\nAdd or update using this row format: | <key> | <absolute-path> | <note> |.\nUse one row per key. If a key already exists, replace that row.\nAlways store absolute paths (no relative paths).\nNormalize key to lowercase kebab-case before writing.\nIf user provides project name, store it in note; otherwise use workspace.\nMissing project mapping flow (required)\n\nWhen project_cwd cannot be resolved, or exact project key is missing in MEMORY.md:\n\nAsk one short question requesting:\nproject key (or confirm inferred key from task prefix)\nabsolute workspace path\noptional project name for note\nValidate path is absolute.\nUpdate or insert the MEMORY.md row immediately.\nConfirm mapping in one short line.\nContinue the original task using the newly resolved project_cwd, then create task worktree cwd and execute.\nBranch naming convention (project key based)\n\nUse this format for worktree branches:\n\ncodex/<project_key>\n\nRules:\n\n<project_key> must be the resolved project key from MEMORY.md.\nNormalize <project_key> to lowercase and kebab-case.\nWorktree-first execution (required default)\n\nAfter resolving project_cwd and project_key, prepare execution workspace:\n\nBuild branch name using this skill's convention:\ncodex/<project_key>\nBuild per-task worktree path:\ndefault root: ${OPENCLAW_WORKTREE_ROOT:-/tmp/openclaw-worktrees}\ndirectory: <worktree_root>/<project_key>\nCreate the worktree from project_cwd before launching engine:\nnew branch: git -C <project_cwd> worktree add -b codex/<project_key> <cwd> HEAD\nexisting branch: git -C <project_cwd> worktree add <cwd> codex/<project_key>\nUse this worktree path as <cwd> for prompt rendering and engine execution.\nPrompt rendering (required template)\n\nDo not forward raw user text directly. Render this template:\n\nIn this template, <cwd> must be the task worktree path (not the primary checkout path).\n\nTask ID: <task or N/A>\nProject Key: <project key or N/A>\nUser Objective: <objective>\nExecution Requirements:\n1) Work only inside <cwd>.\n2) Complete the requested work.\n3) After completion, return a short completion summary.\n\nEngine commands (current policy retained)\n\nAll engine commands must be launched via bash with PTY and explicit workdir.\n\nWhy dangerously-* flags are retained:\n\nThis skill targets non-interactive PTY/background execution.\nPermission prompts can block unattended runs; these flags avoid that blocking behavior.\nThese flags are allowed only after passing the required safety gates above and only in resolved task worktrees.\nIf your environment does not allow these flags, fail fast with a short reason instead of silently falling back.\nClaude Code\nbash pty:true workdir:<cwd> command:\"claude --dangerously-skip-permissions '<rendered_prompt>'\"\n\nCodex CLI\nbash pty:true workdir:<cwd> command:\"codex exec --dangerously-bypass-approvals-and-sandbox '<rendered_prompt>'\"\n\nGemini CLI\nbash pty:true workdir:<cwd> command:\"GEMINI_SANDBOX=false gemini -p '<rendered_prompt>'\"\n\nBash execution interface (required)\n\nUse bash with PTY and optional background mode.\n\nBash parameters\nParameter\tType\tRequired\tPurpose\ncommand\tstring\tyes\tEngine command to run\npty\tboolean\tyes\tMust be true for coding-agent CLIs\nworkdir\tstring\tyes\tPer-task worktree <cwd>\nbackground\tboolean\tno\tRun asynchronously and return session id\ntimeout\tnumber\tno\tHard timeout in seconds\nelevated\tboolean\tno\tHost execution if policy allows\nProcess actions for background sessions\n\nUse these actions as standard controls:\n\nlist: list sessions\npoll: check running/done status\nlog: read incremental output\nwrite: send raw stdin\nsubmit: send stdin + newline\nkill: terminate a session only when required\nExecution patterns (workdir + background + pty)\nOne-shot example\n\nCreate a task worktree, then run inside that worktree:\n\ngit -C <project_cwd> worktree add -b codex/<project_key> /tmp/openclaw-worktrees/<project_key> HEAD\nbash pty:true workdir:/tmp/openclaw-worktrees/<project_key> command:\"codex exec --dangerously-bypass-approvals-and-sandbox '<rendered_prompt>'\"\n\n\nThe Pattern: workdir + background + pty For longer tasks, use background mode with PTY:\n\n# Start agent in task worktree (with PTY!)\nbash pty:true workdir:<cwd> background:true command:\"codex exec --full-auto 'Build a snake game'\"\n# Returns sessionId for tracking\n\n# Monitor progress\nprocess action:log sessionId:XXX\n\n# Check if done\nprocess action:poll sessionId:XXX\n\n# Send input (if agent asks a question)\nprocess action:write sessionId:XXX data:\"y\"\n\n# Submit with Enter (like typing \"yes\" and pressing Enter)\nprocess action:submit sessionId:XXX data:\"yes\"\n\n# Kill if needed\nprocess action:kill sessionId:XXX\n\n\nWhy workdir matters: Agent wakes up in a focused directory, doesn't wander off reading unrelated files (like your soul.md 😅).\n\nIf user input is required mid-run\nprocess action:write sessionId:<id> data:\"y\"\nprocess action:submit sessionId:<id> data:\"yes\"\n\nPR review safety pattern (temp dir/worktree only)\n\nNever run PR review in live OpenClaw folders.\n\n# default: git worktree review (project-key based branch naming)\ngit worktree add -b codex/<project_key> /tmp/<project_key>-review <base_branch>\nbash pty:true workdir:/tmp/<project_key>-review command:\"codex review --base <base_branch>\"\n\n# fallback: temp clone review (only when local checkout is unavailable)\nREVIEW_DIR=$(mktemp -d)\ngit clone <repo> \"$REVIEW_DIR\"\ncd \"$REVIEW_DIR\" && gh pr checkout <pr_number>\nbash pty:true workdir:\"$REVIEW_DIR\" command:\"codex review --base origin/main\"\n\nIssue worktree pattern\ngit worktree add -b codex/<project_key> /tmp/<project_key> main\n\nbash pty:true workdir:/tmp/<project_key> background:true command:\"codex exec --dangerously-bypass-approvals-and-sandbox 'Fix issue #101. Commit after validation.'\"\nbash pty:true workdir:/tmp/<project_key> background:true command:\"codex exec --dangerously-bypass-approvals-and-sandbox 'Fix issue #102. Commit after validation.'\"\n\nprocess action:list\nprocess action:log sessionId:<id>\n\nProgress Updates (Critical)\n\nFor background runs:\n\nWhen you spawn coding agents in the background, keep the user in the loop.\n\nSend 1 short message when you start (what's running + where).\nThen only update again when something changes:\na milestone completes (build finished, tests passed)\nthe agent asks a question / needs input\nyou hit an error or need user action\nthe agent finishes (include what changed + where)\nIf you kill a session, immediately say you killed it and why.\n\nThis prevents the user from seeing only \"Agent failed before reply\" and having no idea what happened.\n\nAuto-notify on completion\n\nFor long-running jobs, append a completion trigger to the rendered prompt:\n\nWhen completely finished, run this command:\nopenclaw system event --text \"Done: <brief summary>\" --mode now\n\n\nExample:\n\nbash pty:true workdir:<cwd> background:true command:\"codex exec --dangerously-bypass-approvals-and-sandbox '<rendered_prompt>\n\nWhen completely finished, run:\nopenclaw system event --text \\\"Done: implemented requested PREQSTATION task\\\" --mode now'\"\n\nOutput policy\n\nReturn only a short completion summary. Success format: completed: <task or N/A> via <engine> at <cwd>\n\nFailure format: failed: <task or N/A> via <engine> at <cwd or N/A> - <short reason>\n\nDo not dump raw stdout/stderr unless user explicitly asks."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/sonim1/preqstation-openclaw",
    "publisherUrl": "https://clawhub.ai/sonim1/preqstation-openclaw",
    "owner": "sonim1",
    "version": "0.1.6",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/preqstation-openclaw",
    "downloadUrl": "https://openagent3.xyz/downloads/preqstation-openclaw",
    "agentUrl": "https://openagent3.xyz/skills/preqstation-openclaw/agent",
    "manifestUrl": "https://openagent3.xyz/skills/preqstation-openclaw/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/preqstation-openclaw/agent.md"
  }
}