# Send Adversarial Prompting to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - 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. ## Suggested prompts ### New install ```text 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. ``` ### Upgrade existing ```text 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. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "adversarial-prompting", "name": "Adversarial Prompting", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/abe238/adversarial-prompting", "canonicalUrl": "https://clawhub.ai/abe238/adversarial-prompting", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/adversarial-prompting", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=adversarial-prompting", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "SKILL.md", "scripts/export_analysis.py" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-23T16:43:11.935Z", "expiresAt": "2026-04-30T16:43:11.935Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard", "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.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/adversarial-prompting" }, "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." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/adversarial-prompting", "downloadUrl": "https://openagent3.xyz/downloads/adversarial-prompting", "agentUrl": "https://openagent3.xyz/skills/adversarial-prompting/agent", "manifestUrl": "https://openagent3.xyz/skills/adversarial-prompting/agent.json", "briefUrl": "https://openagent3.xyz/skills/adversarial-prompting/agent.md" } } ``` ## Documentation ### Adversarial Prompting This skill applies a structured adversarial methodology to problem-solving by generating multiple solutions, rigorously critiquing each for weaknesses, developing fixes, validating those fixes, and consolidating into ranked recommendations. The approach forces deep analysis of failure modes, edge cases, and unintended consequences before committing to a solution. ### When to Use This Skill Use this skill when: Facing complex technical problems requiring thorough analysis (architecture decisions, debugging, performance optimization) Solving strategic or business problems with multiple viable approaches Needing to identify weaknesses in proposed solutions before implementation Requiring validated fixes that address root causes, not symptoms Working on high-stakes decisions where failure modes must be understood Seeking comprehensive analysis with detailed reasoning visible throughout Do not use this skill for: Simple, straightforward problems with obvious solutions Time-sensitive decisions requiring immediate action without analysis Problems where exploration and iteration are more valuable than upfront analysis ### Primary Workflow When invoked, apply the following 7-phase process to the user's problem: Phase 1: Solution Generation Generate 3-7 distinct solution approaches. For each solution: Explain the reasoning behind the approach Describe the core strategy Outline the key steps or components Phase 2: Adversarial Critique For each solution, rigorously identify critical weaknesses. Show thinking while examining: Edge cases and failure modes Security vulnerabilities or risks Performance bottlenecks Scalability limitations Hidden assumptions that could break Resource constraints (time, money, people) Unintended consequences Catastrophic failure scenarios Be creative and thorough in identifying what could go wrong. Phase 3: Fix Development For each identified weakness: Propose a specific fix or mitigation strategy Explain why this fix addresses the root cause Describe how the fix integrates with the original solution Phase 4: Validation Check Review each fix to verify it actually solves the weakness: Confirm the fix addresses the root cause Check for new problems introduced by the fix Flag any remaining concerns or trade-offs Phase 5: Consolidation Synthesize all solutions and validated fixes into comprehensive approaches: Integrate complementary elements from different solutions Eliminate redundancies Show how solutions can be combined for stronger approaches Present the final set of viable options Phase 6: Summary of Options Present all viable options in priority order, ranked by: Feasibility: Can this actually be implemented with available resources? Impact: How well does this solve the problem? Risk Level: What could still go wrong? Resource Requirements: Cost in time, money, and effort For each option, provide a one-paragraph summary highlighting key trade-offs. Phase 7: Final Recommendation State the top recommendation (single option or combination): Clear rationale for why this is the best path Concrete next steps for implementation Key success metrics to track Early warning signs to monitor for problems ### Output Format Present the complete analysis in three sections: Detailed Walkthrough: Show all phases (1-5) with full reasoning visible Summary of Options: Ranked list of viable approaches (Phase 6) Final Recommendation: Top choice with implementation guidance (Phase 7) After presenting the analysis, automatically export the complete output to a markdown file using scripts/export_analysis.py. ### Implementation Notes Show reasoning throughout the process for transparency Be thorough in adversarial critique—surface uncomfortable truths Validate fixes rigorously to avoid creating new problems Consolidation should create stronger solutions, not just list options Final recommendation should be actionable with clear next steps Export results to markdown for future reference and sharing ## Trust - Source: tencent - Verification: Indexed source record - Publisher: abe238 - Version: 1.0.0 ## Source health - Status: healthy - Source download looks usable. - Yavira can redirect you to the upstream package for this source. - Health scope: source - Reason: direct_download_ok - Checked at: 2026-04-23T16:43:11.935Z - Expires at: 2026-04-30T16:43:11.935Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/adversarial-prompting) - [Send to Agent page](https://openagent3.xyz/skills/adversarial-prompting/agent) - [JSON manifest](https://openagent3.xyz/skills/adversarial-prompting/agent.json) - [Markdown brief](https://openagent3.xyz/skills/adversarial-prompting/agent.md) - [Download page](https://openagent3.xyz/downloads/adversarial-prompting)