{ "schemaVersion": "1.0", "item": { "slug": "giga-coding-agent", "name": "Giga Coding Agent", "source": "tencent", "type": "skill", "category": "开发巄具", "sourceUrl": "https://clawhub.ai/branexp/giga-coding-agent", "canonicalUrl": "https://clawhub.ai/branexp/giga-coding-agent", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/giga-coding-agent", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=giga-coding-agent", "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/giga-coding-agent" }, "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/giga-coding-agent", "agentPageUrl": "https://openagent3.xyz/skills/giga-coding-agent/agent", "manifestUrl": "https://openagent3.xyz/skills/giga-coding-agent/agent.json", "briefUrl": "https://openagent3.xyz/skills/giga-coding-agent/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": "Coding Agent (background-first)", "body": "Use bash background mode for non-interactive coding work. For interactive coding sessions, use the tmux skill (always, except very simple one-shot prompts)." }, { "title": "The Pattern: workdir + background", "body": "# Create temp space for chats/scratch work\nSCRATCH=$(mktemp -d)\n\n# Start agent in target directory (\"little box\" - only sees relevant files)\nbash workdir:$SCRATCH background:true command:\"\"\n# Or for project work:\nbash workdir:~/project/folder background:true command:\"\"\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# 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 šŸ˜…)." }, { "title": "Codex CLI", "body": "Model: gpt-5.2-codex is the default (set in ~/.codex/config.toml)" }, { "title": "Building/Creating (use --full-auto or --yolo)", "body": "# --full-auto: sandboxed but auto-approves in workspace\nbash workdir:~/project background:true command:\"codex exec --full-auto \\\"Build a snake game with dark theme\\\"\"\n\n# --yolo: NO sandbox, NO approvals (fastest, most dangerous)\nbash workdir:~/project background:true command:\"codex --yolo \\\"Build a snake game with dark theme\\\"\"\n\n# Note: --yolo is a shortcut for --dangerously-bypass-approvals-and-sandbox" }, { "title": "Reviewing PRs (vanilla, no flags)", "body": "āš ļø CRITICAL: Never review PRs in Clawdbot's own project folder!\n\nEither use the project where the PR is submitted (if it's NOT ~/Projects/clawdbot)\nOr clone to a temp folder first\n\n# Option 1: Review in the actual project (if NOT clawdbot)\nbash workdir:~/Projects/some-other-repo background:true command:\"codex review --base main\"\n\n# Option 2: Clone to temp folder for safe review (REQUIRED for clawdbot PRs!)\nREVIEW_DIR=$(mktemp -d)\ngit clone https://github.com/clawdbot/clawdbot.git $REVIEW_DIR\ncd $REVIEW_DIR && gh pr checkout 130\nbash workdir:$REVIEW_DIR background:true command:\"codex review --base origin/main\"\n# Clean up after: rm -rf $REVIEW_DIR\n\n# Option 3: Use git worktree (keeps main intact)\ngit worktree add /tmp/pr-130-review pr-130-branch\nbash workdir:/tmp/pr-130-review background:true command:\"codex review --base main\"\n\nWhy? Checking out branches in the running Clawdbot repo can break the live instance!" }, { "title": "Batch PR Reviews (parallel army!)", "body": "# Fetch all PR refs first\ngit fetch origin '+refs/pull/*/head:refs/remotes/origin/pr/*'\n\n# Deploy the army - one Codex per PR!\nbash workdir:~/project background:true command:\"codex exec \\\"Review PR #86. git diff origin/main...origin/pr/86\\\"\"\nbash workdir:~/project background:true command:\"codex exec \\\"Review PR #87. git diff origin/main...origin/pr/87\\\"\"\nbash workdir:~/project background:true command:\"codex exec \\\"Review PR #95. git diff origin/main...origin/pr/95\\\"\"\n# ... repeat for all PRs\n\n# Monitor all\nprocess action:list\n\n# Get results and post to GitHub\nprocess action:log sessionId:XXX\ngh pr comment --body \"\"" }, { "title": "Tips for PR Reviews", "body": "Fetch refs first: git fetch origin '+refs/pull/*/head:refs/remotes/origin/pr/*'\nUse git diff: Tell Codex to use git diff origin/main...origin/pr/XX\nDon't checkout: Multiple parallel reviews = don't let them change branches\nPost results: Use gh pr comment to post reviews to GitHub" }, { "title": "Claude Code", "body": "bash workdir:~/project background:true command:\"claude \\\"Your task\\\"\"" }, { "title": "OpenCode", "body": "bash workdir:~/project background:true command:\"opencode run \\\"Your task\\\"\"" }, { "title": "Pi Coding Agent", "body": "# Install: npm install -g @mariozechner/pi-coding-agent\nbash workdir:~/project background:true command:\"pi \\\"Your task\\\"\"" }, { "title": "Pi flags (common)", "body": "--print / -p: non-interactive; runs prompt and exits.\n--provider : pick provider (default: google).\n--model : pick model (default: gemini-2.5-flash).\n--api-key : override API key (defaults to env vars).\n\nExamples:\n\n# Set provider + model, non-interactive\nbash workdir:~/project background:true command:\"pi --provider openai --model gpt-4o-mini -p \\\"Summarize src/\\\"\"" }, { "title": "tmux (interactive sessions)", "body": "Use the tmux skill for interactive coding sessions (always, except very simple one-shot prompts). Prefer bash background mode for non-interactive runs." }, { "title": "Parallel Issue Fixing with git worktrees + tmux", "body": "For fixing multiple issues in parallel, use git worktrees (isolated branches) + tmux sessions:\n\n# 1. Clone repo to temp location\ncd /tmp && git clone git@github.com:user/repo.git repo-worktrees\ncd repo-worktrees\n\n# 2. Create worktrees for each issue (isolated branches!)\ngit worktree add -b fix/issue-78 /tmp/issue-78 main\ngit worktree add -b fix/issue-99 /tmp/issue-99 main\n\n# 3. Set up tmux sessions\nSOCKET=\"${TMPDIR:-/tmp}/codex-fixes.sock\"\ntmux -S \"$SOCKET\" new-session -d -s fix-78\ntmux -S \"$SOCKET\" new-session -d -s fix-99\n\n# 4. Launch Codex in each (after pnpm install!)\ntmux -S \"$SOCKET\" send-keys -t fix-78 \"cd /tmp/issue-78 && pnpm install && codex --yolo 'Fix issue #78: . Commit and push.'\" Enter\ntmux -S \"$SOCKET\" send-keys -t fix-99 \"cd /tmp/issue-99 && pnpm install && codex --yolo 'Fix issue #99: . Commit and push.'\" Enter\n\n# 5. Monitor progress\ntmux -S \"$SOCKET\" capture-pane -p -t fix-78 -S -30\ntmux -S \"$SOCKET\" capture-pane -p -t fix-99 -S -30\n\n# 6. Check if done (prompt returned)\ntmux -S \"$SOCKET\" capture-pane -p -t fix-78 -S -3 | grep -q \"āÆ\" && echo \"Done!\"\n\n# 7. Create PRs after fixes\ncd /tmp/issue-78 && git push -u origin fix/issue-78\ngh pr create --repo user/repo --head fix/issue-78 --title \"fix: ...\" --body \"...\"\n\n# 8. Cleanup\ntmux -S \"$SOCKET\" kill-server\ngit worktree remove /tmp/issue-78\ngit worktree remove /tmp/issue-99\n\nWhy worktrees? Each Codex works in isolated branch, no conflicts. Can run 5+ parallel fixes!\n\nWhy tmux over bash background? Codex is interactive ā€” needs TTY for proper output. tmux provides persistent sessions with full history capture." }, { "title": "āš ļø Rules", "body": "Respect tool choice ā€” if user asks for Codex, use Codex. NEVER offer to build it yourself!\nBe patient ā€” don't kill sessions because they're \"slow\"\nMonitor with process:log ā€” check progress without interfering\n--full-auto for building ā€” auto-approves changes\nvanilla for reviewing ā€” no special flags needed\nParallel is OK ā€” run many Codex processes at once for batch work\nNEVER start Codex in ~/clawd/ ā€” it'll read your soul docs and get weird ideas about the org chart! Use the target project dir or /tmp for blank slate chats\nNEVER checkout branches in ~/Projects/clawdbot/ ā€” that's the LIVE Clawdbot instance! Clone to /tmp or use git worktree for PR reviews" }, { "title": "PR Template (The Razor Standard)", "body": "When submitting PRs to external repos, use this format for quality & maintainer-friendliness:\n\n## Original Prompt\n[Exact request/problem statement]\n\n## What this does\n[High-level description]\n\n**Features:**\n- [Key feature 1]\n- [Key feature 2]\n\n**Example usage:**\n```bash\n# Example\ncommand example\n```\n\n## Feature intent (maintainer-friendly)\n[Why useful, how it fits, workflows it enables]\n\n## Prompt history (timestamped)\n- YYYY-MM-DD HH:MM UTC: [Step 1]\n- YYYY-MM-DD HH:MM UTC: [Step 2]\n\n## How I tested\n**Manual verification:**\n1. [Test step] - Output: `[result]`\n2. [Test step] - Result: [result]\n\n**Files tested:**\n- [Detail]\n- [Edge cases]\n\n## Session logs (implementation)\n- [What was researched]\n- [What was discovered]\n- [Time spent]\n\n## Implementation details\n**New files:**\n- `path/file.ts` - [description]\n\n**Modified files:**\n- `path/file.ts` - [change]\n\n**Technical notes:**\n- [Detail 1]\n- [Detail 2]\n\n---\n*Submitted by Razor šŸ„· - Mariano's AI agent*\n\nKey principles:\n\nHuman-written description (no AI slop)\nFeature intent for maintainers\nTimestamped prompt history\nSession logs if using Codex/agent\n\nExample: https://github.com/steipete/bird/pull/22" } ], "body": "Coding Agent (background-first)\n\nUse bash background mode for non-interactive coding work. For interactive coding sessions, use the tmux skill (always, except very simple one-shot prompts).\n\nThe Pattern: workdir + background\n# Create temp space for chats/scratch work\nSCRATCH=$(mktemp -d)\n\n# Start agent in target directory (\"little box\" - only sees relevant files)\nbash workdir:$SCRATCH background:true command:\"\"\n# Or for project work:\nbash workdir:~/project/folder background:true command:\"\"\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# 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 šŸ˜…).\n\nCodex CLI\n\nModel: gpt-5.2-codex is the default (set in ~/.codex/config.toml)\n\nBuilding/Creating (use --full-auto or --yolo)\n# --full-auto: sandboxed but auto-approves in workspace\nbash workdir:~/project background:true command:\"codex exec --full-auto \\\"Build a snake game with dark theme\\\"\"\n\n# --yolo: NO sandbox, NO approvals (fastest, most dangerous)\nbash workdir:~/project background:true command:\"codex --yolo \\\"Build a snake game with dark theme\\\"\"\n\n# Note: --yolo is a shortcut for --dangerously-bypass-approvals-and-sandbox\n\nReviewing PRs (vanilla, no flags)\n\nāš ļø CRITICAL: Never review PRs in Clawdbot's own project folder!\n\nEither use the project where the PR is submitted (if it's NOT ~/Projects/clawdbot)\nOr clone to a temp folder first\n# Option 1: Review in the actual project (if NOT clawdbot)\nbash workdir:~/Projects/some-other-repo background:true command:\"codex review --base main\"\n\n# Option 2: Clone to temp folder for safe review (REQUIRED for clawdbot PRs!)\nREVIEW_DIR=$(mktemp -d)\ngit clone https://github.com/clawdbot/clawdbot.git $REVIEW_DIR\ncd $REVIEW_DIR && gh pr checkout 130\nbash workdir:$REVIEW_DIR background:true command:\"codex review --base origin/main\"\n# Clean up after: rm -rf $REVIEW_DIR\n\n# Option 3: Use git worktree (keeps main intact)\ngit worktree add /tmp/pr-130-review pr-130-branch\nbash workdir:/tmp/pr-130-review background:true command:\"codex review --base main\"\n\n\nWhy? Checking out branches in the running Clawdbot repo can break the live instance!\n\nBatch PR Reviews (parallel army!)\n# Fetch all PR refs first\ngit fetch origin '+refs/pull/*/head:refs/remotes/origin/pr/*'\n\n# Deploy the army - one Codex per PR!\nbash workdir:~/project background:true command:\"codex exec \\\"Review PR #86. git diff origin/main...origin/pr/86\\\"\"\nbash workdir:~/project background:true command:\"codex exec \\\"Review PR #87. git diff origin/main...origin/pr/87\\\"\"\nbash workdir:~/project background:true command:\"codex exec \\\"Review PR #95. git diff origin/main...origin/pr/95\\\"\"\n# ... repeat for all PRs\n\n# Monitor all\nprocess action:list\n\n# Get results and post to GitHub\nprocess action:log sessionId:XXX\ngh pr comment --body \"\"\n\nTips for PR Reviews\nFetch refs first: git fetch origin '+refs/pull/*/head:refs/remotes/origin/pr/*'\nUse git diff: Tell Codex to use git diff origin/main...origin/pr/XX\nDon't checkout: Multiple parallel reviews = don't let them change branches\nPost results: Use gh pr comment to post reviews to GitHub\nClaude Code\nbash workdir:~/project background:true command:\"claude \\\"Your task\\\"\"\n\nOpenCode\nbash workdir:~/project background:true command:\"opencode run \\\"Your task\\\"\"\n\nPi Coding Agent\n# Install: npm install -g @mariozechner/pi-coding-agent\nbash workdir:~/project background:true command:\"pi \\\"Your task\\\"\"\n\nPi flags (common)\n--print / -p: non-interactive; runs prompt and exits.\n--provider : pick provider (default: google).\n--model : pick model (default: gemini-2.5-flash).\n--api-key : override API key (defaults to env vars).\n\nExamples:\n\n# Set provider + model, non-interactive\nbash workdir:~/project background:true command:\"pi --provider openai --model gpt-4o-mini -p \\\"Summarize src/\\\"\"\n\ntmux (interactive sessions)\n\nUse the tmux skill for interactive coding sessions (always, except very simple one-shot prompts). Prefer bash background mode for non-interactive runs.\n\nParallel Issue Fixing with git worktrees + tmux\n\nFor fixing multiple issues in parallel, use git worktrees (isolated branches) + tmux sessions:\n\n# 1. Clone repo to temp location\ncd /tmp && git clone git@github.com:user/repo.git repo-worktrees\ncd repo-worktrees\n\n# 2. Create worktrees for each issue (isolated branches!)\ngit worktree add -b fix/issue-78 /tmp/issue-78 main\ngit worktree add -b fix/issue-99 /tmp/issue-99 main\n\n# 3. Set up tmux sessions\nSOCKET=\"${TMPDIR:-/tmp}/codex-fixes.sock\"\ntmux -S \"$SOCKET\" new-session -d -s fix-78\ntmux -S \"$SOCKET\" new-session -d -s fix-99\n\n# 4. Launch Codex in each (after pnpm install!)\ntmux -S \"$SOCKET\" send-keys -t fix-78 \"cd /tmp/issue-78 && pnpm install && codex --yolo 'Fix issue #78: . Commit and push.'\" Enter\ntmux -S \"$SOCKET\" send-keys -t fix-99 \"cd /tmp/issue-99 && pnpm install && codex --yolo 'Fix issue #99: . Commit and push.'\" Enter\n\n# 5. Monitor progress\ntmux -S \"$SOCKET\" capture-pane -p -t fix-78 -S -30\ntmux -S \"$SOCKET\" capture-pane -p -t fix-99 -S -30\n\n# 6. Check if done (prompt returned)\ntmux -S \"$SOCKET\" capture-pane -p -t fix-78 -S -3 | grep -q \"āÆ\" && echo \"Done!\"\n\n# 7. Create PRs after fixes\ncd /tmp/issue-78 && git push -u origin fix/issue-78\ngh pr create --repo user/repo --head fix/issue-78 --title \"fix: ...\" --body \"...\"\n\n# 8. Cleanup\ntmux -S \"$SOCKET\" kill-server\ngit worktree remove /tmp/issue-78\ngit worktree remove /tmp/issue-99\n\n\nWhy worktrees? Each Codex works in isolated branch, no conflicts. Can run 5+ parallel fixes!\n\nWhy tmux over bash background? Codex is interactive ā€” needs TTY for proper output. tmux provides persistent sessions with full history capture.\n\nāš ļø Rules\nRespect tool choice ā€” if user asks for Codex, use Codex. NEVER offer to build it yourself!\nBe patient ā€” don't kill sessions because they're \"slow\"\nMonitor with process:log ā€” check progress without interfering\n--full-auto for building ā€” auto-approves changes\nvanilla for reviewing ā€” no special flags needed\nParallel is OK ā€” run many Codex processes at once for batch work\nNEVER start Codex in ~/clawd/ ā€” it'll read your soul docs and get weird ideas about the org chart! Use the target project dir or /tmp for blank slate chats\nNEVER checkout branches in ~/Projects/clawdbot/ ā€” that's the LIVE Clawdbot instance! Clone to /tmp or use git worktree for PR reviews\nPR Template (The Razor Standard)\n\nWhen submitting PRs to external repos, use this format for quality & maintainer-friendliness:\n\n## Original Prompt\n[Exact request/problem statement]\n\n## What this does\n[High-level description]\n\n**Features:**\n- [Key feature 1]\n- [Key feature 2]\n\n**Example usage:**\n```bash\n# Example\ncommand example\n```\n\n## Feature intent (maintainer-friendly)\n[Why useful, how it fits, workflows it enables]\n\n## Prompt history (timestamped)\n- YYYY-MM-DD HH:MM UTC: [Step 1]\n- YYYY-MM-DD HH:MM UTC: [Step 2]\n\n## How I tested\n**Manual verification:**\n1. [Test step] - Output: `[result]`\n2. [Test step] - Result: [result]\n\n**Files tested:**\n- [Detail]\n- [Edge cases]\n\n## Session logs (implementation)\n- [What was researched]\n- [What was discovered]\n- [Time spent]\n\n## Implementation details\n**New files:**\n- `path/file.ts` - [description]\n\n**Modified files:**\n- `path/file.ts` - [change]\n\n**Technical notes:**\n- [Detail 1]\n- [Detail 2]\n\n---\n*Submitted by Razor šŸ„· - Mariano's AI agent*\n\n\nKey principles:\n\nHuman-written description (no AI slop)\nFeature intent for maintainers\nTimestamped prompt history\nSession logs if using Codex/agent\n\nExample: https://github.com/steipete/bird/pull/22" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/branexp/giga-coding-agent", "publisherUrl": "https://clawhub.ai/branexp/giga-coding-agent", "owner": "branexp", "version": "1.0.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/giga-coding-agent", "downloadUrl": "https://openagent3.xyz/downloads/giga-coding-agent", "agentUrl": "https://openagent3.xyz/skills/giga-coding-agent/agent", "manifestUrl": "https://openagent3.xyz/skills/giga-coding-agent/agent.json", "briefUrl": "https://openagent3.xyz/skills/giga-coding-agent/agent.md" } }