adversarial-plan-review
Fresh adversarial review of implementation plans before execution. Dispatches a read-only agent to challenge feasibility, find gaps, and surface unstated assumptions. Bounded to 2 review passes. Use after writing a plan and before starting implementation.
What this skill does
# Adversarial Plan Review After writing an implementation plan, challenge it with a fresh perspective before committing to execution. This catches feasibility gaps, missing edge cases, and unstated assumptions that the planner (you) may be anchored to. ## When to Use - After completing a plan (via `superpowers:writing-plans` or manual planning) - Before executing any multi-step implementation - When a plan involves unfamiliar technology, complex integrations, or architectural changes ## Protocol ### Step 1 — Save the Plan Ensure the plan exists as a readable artifact (file or task description). The reviewer needs the full plan, not a summary. ### Step 2 — Dispatch Adversarial Reviewer Launch a **read-only** Agent with this framing: ``` You are a fresh reviewer seeing this plan for the first time. You have NO ownership of it and NO anchoring bias. Review the plan at [path/location] for: 1. **Feasibility gaps**: Are there steps that assume something unverified? (API exists, library supports X, performance will be acceptable) 2. **Missing edge cases**: What happens when inputs are empty, huge, concurrent, or malformed? What about error paths? 3. **Unstated assumptions**: What does the plan take for granted that should be explicitly verified or decided? 4. **Ordering risks**: Are there steps that depend on earlier steps but don't say so? Could parallel execution cause conflicts? 5. **Scope creep signals**: Are there steps that go beyond the original requirement? For each concern: - State it as a specific question, not vague worry - Reference the specific plan step - Rate: BLOCKING (must resolve before execution) or ADVISORY (worth noting) If the plan is solid, say so explicitly — do NOT invent concerns to appear thorough. ``` ### Step 3 — Evaluate Findings | Finding type | Action | |-------------|--------| | No concerns (plan is solid) | Proceed to execution | | ADVISORY only | Note them, proceed to execution | | BLOCKING concerns (pass 1) | Revise the plan to address them, then run one more review | | BLOCKING concerns (pass 2) | Present unresolved concerns to user for decision | ### Hard Limits - **Maximum 2 review passes.** Do not loop further. If 2 passes surface persistent concerns, the user decides. - **Reviewer is read-only.** It challenges the plan but does not rewrite it. You (the planner) own revisions. - **No rubber stamps.** If the reviewer returns zero concerns on a complex plan, that is a signal to check whether it actually read the plan — not a green light to skip scrutiny. ## Anti-Patterns - Skipping review for "simple" plans (the simple ones are where hidden assumptions live) - Letting the reviewer rewrite the plan (it advises, you decide) - Dismissing BLOCKING concerns without addressing them - Running more than 2 passes (diminishing returns, increasing anchoring) - Reviewing your own plan without a fresh agent (you are anchored to your own thinking)
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.