{
  "schemaVersion": "1.0",
  "item": {
    "slug": "red-team",
    "name": "Red Team",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/retrodigio/red-team",
    "canonicalUrl": "https://clawhub.ai/retrodigio/red-team",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/red-team",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=red-team",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "README.md",
      "SKILL.md",
      "references/personas.md",
      "scripts/red-team.py"
    ],
    "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": "2026-05-07T17:22:31.273Z",
      "expiresAt": "2026-05-14T17:22:31.273Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
        "contentDisposition": "attachment; filename=\"afrexai-annual-report-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/red-team"
    },
    "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/red-team",
    "agentPageUrl": "https://openagent3.xyz/skills/red-team/agent",
    "manifestUrl": "https://openagent3.xyz/skills/red-team/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/red-team/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": "Red Team — Adversarial Debate Engine",
        "body": "Stress-test any decision by having AI agents with conflicting worldviews debate it."
      },
      {
        "title": "Prerequisites",
        "body": "One of these coding agent CLIs (uses your existing subscription — no API key needed):\n\nClaude Code (default): claude — npm i -g @anthropic-ai/claude-code\nCodex: codex — npm i -g @openai/codex\nGemini: gemini — npm i -g @google/gemini-cli\n\nNo Python dependencies beyond the standard library."
      },
      {
        "title": "Quick Start",
        "body": "# Basic 3-persona debate (uses Max subscription via claude CLI)\npython3 ~/.openclaw/skills/red-team/scripts/red-team.py \\\n  --question \"Should we do X?\" \\\n  --personas \"bull,bear,operator\"\n\n# Full debate with context and output file\npython3 ~/.openclaw/skills/red-team/scripts/red-team.py \\\n  -q \"Should we invest $50k in this deal?\" \\\n  -p \"bull,bear,cash-flow,local-realist\" \\\n  -r 3 \\\n  -c /path/to/deal-data.md \\\n  -o /tmp/red-team-result.md\n\n# Use a different model\npython3 ~/.openclaw/skills/red-team/scripts/red-team.py \\\n  -q \"Should we launch this product?\" \\\n  -p \"bull,customer,operator\" \\\n  -m opus\n\n# List all available personas\npython3 ~/.openclaw/skills/red-team/scripts/red-team.py --list-personas"
      },
      {
        "title": "How to Use (as OpenClaw Agent)",
        "body": "When the user asks you to \"red team\" something, \"stress test\" an idea, play \"devil's advocate\", or asks \"what could go wrong\":\n\nIdentify the question/decision from the user's message\nChoose appropriate personas (default: bull,bear,operator — adjust based on domain)\nRun the script and save output\nSummarize the key findings to the user, share the full report if requested\n\nPersona selection guide:\n\nInvestment/financial decisions → bull, bear, cash-flow, economist\nProduct/startup ideas → bull, customer, operator, technologist\nLegal/compliance questions → regulator, bear, operator\nStrategy/direction → contrarian, economist, historian, bull\nGeneral \"should we do X?\" → bull, bear, operator (good default)"
      },
      {
        "title": "Available Personas",
        "body": "KeyNameWorldviewbullThe BullOptimistic, opportunity-focusedbearThe BearRisk-averse, capital preservationcontrarianThe ContrarianOppositional, consensus-challengingoperatorThe OperatorExecution-focused pragmatisteconomistThe EconomistMacro trends, opportunity costlocal-realistThe Local RealistGround truth, local specificscash-flowThe Cash Flow AnalystIncome, carrying costs, IRRregulatorThe RegulatorCompliance, legal risktechnologistThe TechnologistAutomation, scalabilitycustomerThe CustomerEnd-user demand, willingness to payethicistThe EthicistMoral implications, stakeholder impacthistorianThe HistorianHistorical patterns, precedent"
      },
      {
        "title": "Custom Personas",
        "body": "Create a JSON file:\n\n{\n  \"my-persona\": {\n    \"name\": \"The Skeptic\",\n    \"description\": \"Questions everything, trusts nothing\",\n    \"system\": \"You are The Skeptic — you question every assumption...\"\n  }\n}\n\nUse with --custom-personas /path/to/file.json. Custom personas merge with built-ins."
      },
      {
        "title": "CLI Options",
        "body": "FlagDefaultDescription--question, -qrequiredThe question to debate--personas, -pbull,bear,operatorComma-separated persona keys--rounds, -r2Number of critique rounds--output, -ostdoutOutput file path--context-file, -cnoneAdditional context file--custom-personasnoneCustom personas JSON--model, -msonnetModel alias (sonnet, opus, haiku, gpt-4o, etc.)--backend, -bclaudeCLI backend: claude, codex, or gemini--list-personas—List personas and exit"
      },
      {
        "title": "Output Structure",
        "body": "The output is a markdown document with:\n\nInitial Proposals — Each agent's independent take\nCritique Rounds — Agents critique each other\nRefinement — Agents update positions based on critiques\nConviction Scores — Each agent scores all positions (0-100)\nSynthesis & Decision Brief — Neutral agent produces:\n\nExecutive summary\nConsensus points\nKey disagreements\nRisk matrix\nConviction score summary\nSynthesized recommendation\nNext steps"
      },
      {
        "title": "When to Use",
        "body": "✅ Good for: Important decisions, investment analysis, product strategy, \"go/no-go\" calls, pre-mortems, challenging groupthink\n\n❌ Not for: Simple factual questions, time-sensitive emergencies, decisions already made, emotional/personal choices"
      },
      {
        "title": "Integration Tips",
        "body": "Save output to memory files for future reference\nCreate BEADS tasks from the \"Next Steps\" section\nFeed context files from Obsidian or project docs\nRe-run with different personas for different perspectives\nUse --rounds 1 for quick takes, --rounds 3 for deep analysis"
      }
    ],
    "body": "Red Team — Adversarial Debate Engine\n\nStress-test any decision by having AI agents with conflicting worldviews debate it.\n\nPrerequisites\n\nOne of these coding agent CLIs (uses your existing subscription — no API key needed):\n\nClaude Code (default): claude — npm i -g @anthropic-ai/claude-code\nCodex: codex — npm i -g @openai/codex\nGemini: gemini — npm i -g @google/gemini-cli\n\nNo Python dependencies beyond the standard library.\n\nQuick Start\n# Basic 3-persona debate (uses Max subscription via claude CLI)\npython3 ~/.openclaw/skills/red-team/scripts/red-team.py \\\n  --question \"Should we do X?\" \\\n  --personas \"bull,bear,operator\"\n\n# Full debate with context and output file\npython3 ~/.openclaw/skills/red-team/scripts/red-team.py \\\n  -q \"Should we invest $50k in this deal?\" \\\n  -p \"bull,bear,cash-flow,local-realist\" \\\n  -r 3 \\\n  -c /path/to/deal-data.md \\\n  -o /tmp/red-team-result.md\n\n# Use a different model\npython3 ~/.openclaw/skills/red-team/scripts/red-team.py \\\n  -q \"Should we launch this product?\" \\\n  -p \"bull,customer,operator\" \\\n  -m opus\n\n# List all available personas\npython3 ~/.openclaw/skills/red-team/scripts/red-team.py --list-personas\n\nHow to Use (as OpenClaw Agent)\n\nWhen the user asks you to \"red team\" something, \"stress test\" an idea, play \"devil's advocate\", or asks \"what could go wrong\":\n\nIdentify the question/decision from the user's message\nChoose appropriate personas (default: bull,bear,operator — adjust based on domain)\nRun the script and save output\nSummarize the key findings to the user, share the full report if requested\n\nPersona selection guide:\n\nInvestment/financial decisions → bull, bear, cash-flow, economist\nProduct/startup ideas → bull, customer, operator, technologist\nLegal/compliance questions → regulator, bear, operator\nStrategy/direction → contrarian, economist, historian, bull\nGeneral \"should we do X?\" → bull, bear, operator (good default)\nAvailable Personas\nKey\tName\tWorldview\nbull\tThe Bull\tOptimistic, opportunity-focused\nbear\tThe Bear\tRisk-averse, capital preservation\ncontrarian\tThe Contrarian\tOppositional, consensus-challenging\noperator\tThe Operator\tExecution-focused pragmatist\neconomist\tThe Economist\tMacro trends, opportunity cost\nlocal-realist\tThe Local Realist\tGround truth, local specifics\ncash-flow\tThe Cash Flow Analyst\tIncome, carrying costs, IRR\nregulator\tThe Regulator\tCompliance, legal risk\ntechnologist\tThe Technologist\tAutomation, scalability\ncustomer\tThe Customer\tEnd-user demand, willingness to pay\nethicist\tThe Ethicist\tMoral implications, stakeholder impact\nhistorian\tThe Historian\tHistorical patterns, precedent\nCustom Personas\n\nCreate a JSON file:\n\n{\n  \"my-persona\": {\n    \"name\": \"The Skeptic\",\n    \"description\": \"Questions everything, trusts nothing\",\n    \"system\": \"You are The Skeptic — you question every assumption...\"\n  }\n}\n\n\nUse with --custom-personas /path/to/file.json. Custom personas merge with built-ins.\n\nCLI Options\nFlag\tDefault\tDescription\n--question, -q\trequired\tThe question to debate\n--personas, -p\tbull,bear,operator\tComma-separated persona keys\n--rounds, -r\t2\tNumber of critique rounds\n--output, -o\tstdout\tOutput file path\n--context-file, -c\tnone\tAdditional context file\n--custom-personas\tnone\tCustom personas JSON\n--model, -m\tsonnet\tModel alias (sonnet, opus, haiku, gpt-4o, etc.)\n--backend, -b\tclaude\tCLI backend: claude, codex, or gemini\n--list-personas\t—\tList personas and exit\nOutput Structure\n\nThe output is a markdown document with:\n\nInitial Proposals — Each agent's independent take\nCritique Rounds — Agents critique each other\nRefinement — Agents update positions based on critiques\nConviction Scores — Each agent scores all positions (0-100)\nSynthesis & Decision Brief — Neutral agent produces:\nExecutive summary\nConsensus points\nKey disagreements\nRisk matrix\nConviction score summary\nSynthesized recommendation\nNext steps\nWhen to Use\n\n✅ Good for: Important decisions, investment analysis, product strategy, \"go/no-go\" calls, pre-mortems, challenging groupthink\n\n❌ Not for: Simple factual questions, time-sensitive emergencies, decisions already made, emotional/personal choices\n\nIntegration Tips\nSave output to memory files for future reference\nCreate BEADS tasks from the \"Next Steps\" section\nFeed context files from Obsidian or project docs\nRe-run with different personas for different perspectives\nUse --rounds 1 for quick takes, --rounds 3 for deep analysis"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/retrodigio/red-team",
    "publisherUrl": "https://clawhub.ai/retrodigio/red-team",
    "owner": "retrodigio",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/red-team",
    "downloadUrl": "https://openagent3.xyz/downloads/red-team",
    "agentUrl": "https://openagent3.xyz/skills/red-team/agent",
    "manifestUrl": "https://openagent3.xyz/skills/red-team/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/red-team/agent.md"
  }
}