{"id":"b4662567-9b65-4cb5-a500-db752f9e0446","shortId":"8LGMME","kind":"skill","title":"utility-pm-critic","tagline":"Run adversarial review on a PM artifact via the pm-critic sub-agent. Dispatches natively on Claude Code with the pm-skills plugin (invokes @agent-pm-critic); on non-Claude clients (Codex CLI, Cursor, Windsurf, Copilot, Gemini CLI) reads subagents/pm-critic.md and executes the sys","description":"<!-- PM-Skills | https://github.com/product-on-purpose/pm-skills | Apache 2.0 -->\n# PM Critic (Dispatch Skill)\n\nThis skill is a cross-client dispatch wrapper for the `pm-critic` sub-agent. It exists so that users on non-Claude clients can run adversarial review with the same intent as Claude Code users, without depending on native plugin sub-agent infrastructure.\n\nPer master plan D11 (amended) + D30, sub-agents are a Claude Code plugin feature. Non-Claude clients (Codex CLI, Cursor, Windsurf, Copilot, Gemini CLI) cannot natively load `subagents/pm-critic.md`. This skill bridges the gap.\n\n## When to Use\n\n- You want adversarial review of a PM artifact (PRD, OKR set, persona, lean canvas, meeting recap, interview synthesis, problem statement, hypothesis, edge case catalog, retrospective, etc.)\n- You are running on a non-Claude AI client and the `pm-critic` sub-agent is not natively available\n- You are running on Claude Code and prefer skill-invocation semantics over sub-agent semantics (e.g., for consistency across a multi-step workflow that mixes skills and sub-agent intents)\n\n## When NOT to Use\n\n- You want a structural lint / repo-level audit -> use `utility-pm-skill-auditor` (audits skills + repo state) instead\n- You want to author an artifact (this skill only reviews) -> use the appropriate phase skill (deliver-prd, foundation-okr-writer, etc.)\n- You want code review -> use a code-review-specific tool (this skill is for PM artifacts)\n- You want to enforce style rules like em-dash sweep -> that is `pm-release-conductor`'s G0 gate, not this skill\n\n## Instructions\n\n**Runtime detection step.** Determine which AI client is invoking this skill.\n\n### If you are running in Claude Code with the pm-skills plugin installed\n\nInvoke `@agent-pm-critic` on the target artifact. Pass the artifact path as argument (or the most recent artifact in session context if no argument is provided). Return the sub-agent's findings to the user. No further action needed from this skill - the sub-agent handles the review natively in its own context window.\n\n### If you are running in any other AI client\n\nCodex CLI, Cursor, Windsurf, Copilot, Gemini CLI, ChatGPT, or any other client without native pm-skills plugin sub-agent support:\n\n1. Read the canonical sub-agent definition at `subagents/pm-critic.md`\n2. Execute the system prompt body in that file as your operating instructions for this turn\n3. Read the target PM artifact specified by the user (path from `$ARGUMENTS`, or most recently produced artifact in session)\n4. Read the canonical standards docs referenced by the pm-critic system prompt for the artifact type (e.g., `skills/foundation-okr-writer/SKILL.md` for OKR sets, `skills/deliver-prd/SKILL.md` for PRDs)\n5. Produce findings graded P0/P1/P2/P3 with concrete fix suggestions, formatted per the output structure documented in `docs/guides/adversarial-review.md`\n6. End the output with the layered structure per master plan D26:\n   - Section (1): full human-readable findings (the P0/P1/P2/P3 report)\n   - Section (2): `## Status Summary` in prose, summarizing what was found and what the user should do next\n   - Section (3): `## Status` YAML block with machine-readable fields\n\n## Output Format\n\nSee `references/TEMPLATE.md` for the canonical output structure (with the layered Status envelope per D26). See `references/EXAMPLE.md` for a worked example showing a real cross-client dispatch run against a PRD.\n\n## Composition\n\n- **Skills:** This dispatch skill composes with all PM-artifact-producing skills (deliver-prd, foundation-okr-writer, foundation-meeting-recap, foundation-persona, foundation-lean-canvas, discover-interview-synthesis, etc.). Run any of those, then run this skill on the produced artifact.\n- **Sub-agents:** On Claude Code, this skill dispatches to `pm-critic` sub-agent. On non-Claude, this skill IS the inline execution; no further dispatch.\n- **Workflows:** Once `pm-workflow-orchestrator` (v2.17) ships, workflows can invoke this skill at quality-gate steps for cross-client compatibility.\n\n## Cross-Client Notes\n\nSee [Sub-Agent Compatibility Matrix](../../docs/reference/sub-agent-compatibility.md) for the canonical cross-client status of all 4 sub-agents + dispatch skills. Summary for this skill as of v2.16.0: PRODUCTION on Claude Code + Codex CLI; EXPERIMENTAL on Cursor / Windsurf / Copilot CLI / Gemini CLI. Maintainers wanting cross-client confidence on a non-Claude client can re-run the [`maintainer-gate-testing-codex.md`](../../docs/internal/release-plans/v2.16.0/maintainer-gate-testing-codex.md) harness on that client.\n\nThe \"read and execute inline\" pattern depends on the AI being able to:\n\n1. Read a file path provided as a reference (most AI clients support this)\n2. Treat that file's content as operating instructions for the current turn (most AI clients support this for SKILL.md-style instructions)\n3. Read additional referenced standards docs at invocation time (all major AI clients support this)\n\nIf any of these are unreliable on a given client, that client cannot use this dispatch skill effectively.\n\n## Reference Files\n\n- Canonical sub-agent definition: [`subagents/pm-critic.md`](../../subagents/pm-critic.md)\n- Behavioral spec: [`docs/internal/release-plans/v2.16.0/spec_pm-critic.md`](../../docs/internal/release-plans/v2.16.0/spec_pm-critic.md)\n- User-facing guide: [`docs/guides/adversarial-review.md`](../../docs/guides/adversarial-review.md)\n- Runtime components catalog: [`docs/reference/runtime-components.md`](../../docs/reference/runtime-components.md)\n- Output template: `references/TEMPLATE.md`\n- Worked example: `references/EXAMPLE.md`","tags":["utility","critic","skills","product-on-purpose","agent-skills","ai-skills","claude-code","claude-desktop","design-sprint","foundation-sprint","openskills","product-management"],"capabilities":["skill","source-product-on-purpose","skill-utility-pm-critic","topic-agent-skills","topic-ai-skills","topic-claude-code","topic-claude-desktop","topic-design-sprint","topic-foundation-sprint","topic-openskills","topic-product-management","topic-skill-families","topic-triple-diamond"],"categories":["pm-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/product-on-purpose/pm-skills/utility-pm-critic","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add product-on-purpose/pm-skills","source_repo":"https://github.com/product-on-purpose/pm-skills","install_from":"skills.sh"}},"qualityScore":"0.560","qualityRationale":"deterministic score 0.56 from registry signals: · indexed on github topic:agent-skills · 221 github stars · SKILL.md body (5,634 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T18:55:01.699Z","embedding":null,"createdAt":"2026-05-17T18:54:31.380Z","updatedAt":"2026-05-18T18:55:01.699Z","lastSeenAt":"2026-05-18T18:55:01.699Z","tsv":"'/../docs/guides/adversarial-review.md':869 '/../docs/internal/release-plans/v2.16.0/maintainer-gate-testing-codex.md':764 '/../docs/internal/release-plans/v2.16.0/spec_pm-critic.md':863 '/../docs/reference/runtime-components.md':874 '/../docs/reference/sub-agent-compatibility.md':709 '/../subagents/pm-critic.md':859 '1':428,530,782 '2':438,540,796 '3':454,557,818 '4':474,719 '5':500 '6':517 'abl':780 'across':212 'action':379 'addit':820 'adversari':6,87,146 'agent':19,33,74,104,114,187,207,224,341,371,387,426,434,649,662,706,722,856 'agent-pm-crit':32,340 'ai':178,319,404,778,792,810,829 'amend':110 'appropri':262 'argument':353,364,466 'artifact':11,151,255,289,347,350,358,459,471,490,609,646 'audit':238,245 'auditor':244 'author':253 'avail':191 'behavior':860 'block':560 'bodi':443 'bridg':138 'cannot':132,845 'canon':431,477,572,712,853 'canva':157,629 'case':166 'catalog':167,872 'chatgpt':413 'claud':23,39,83,94,117,123,177,196,330,651,666,734,756 'cli':42,47,126,131,407,412,737,743,745 'client':40,64,84,124,179,320,405,417,593,697,701,715,750,757,768,793,811,830,842,844 'code':24,95,118,197,275,280,331,652,735 'code-review-specif':279 'codex':41,125,406,736 'compat':698,707 'compon':871 'compos':604 'composit':599 'concret':506 'conductor':306 'confid':751 'consist':211 'content':801 'context':361,395 'copilot':45,129,410,742 'critic':4,16,35,55,71,184,343,485,659 'cross':63,592,696,700,714,749 'cross-client':62,591,695,699,713,748 'current':807 'cursor':43,127,408,740 'd11':109 'd26':528,581 'd30':111 'dash':299 'definit':435,857 'deliv':266,613 'deliver-prd':265,612 'depend':98,775 'detect':315 'determin':317 'discov':631 'discover-interview-synthesi':630 'dispatch':20,56,65,594,602,655,675,723,848 'doc':479,823 'docs/guides/adversarial-review.md':516,868 'docs/internal/release-plans/v2.16.0/spec_pm-critic.md':862 'docs/reference/runtime-components.md':873 'document':514 'e.g':209,492 'edg':165 'effect':850 'em':298 'em-dash':297 'end':518 'enforc':293 'envelop':579 'etc':169,272,634 'exampl':587,879 'execut':51,439,672,772 'exist':76 'experiment':738 'face':866 'featur':120 'field':565 'file':446,785,799,852 'find':373,502,535 'fix':507 'format':509,567 'found':548 'foundat':269,616,620,624,627 'foundation-lean-canva':626 'foundation-meeting-recap':619 'foundation-okr-writ':268,615 'foundation-persona':623 'full':531 'g0':308 'gap':140 'gate':309,692 'gemini':46,130,411,744 'given':841 'grade':503 'guid':867 'handl':388 'har':765 'human':533 'human-read':532 'hypothesi':164 'infrastructur':105 'inlin':671,773 'instal':338 'instead':249 'instruct':313,450,804,817 'intent':92,225 'interview':160,632 'invoc':202,825 'invok':31,322,339,686 'layer':523,577 'lean':156,628 'level':237 'like':296 'lint':234 'load':134 'machin':563 'machine-read':562 'maintain':746 'maintainer-gate-testing-codex.md':763 'major':828 'master':107,526 'matrix':708 'meet':158,621 'mix':219 'multi':215 'multi-step':214 'nativ':21,100,133,190,391,419 'need':380 'next':555 'non':38,82,122,176,665,755 'non-claud':37,81,121,175,664,754 'note':702 'okr':153,270,495,617 'oper':449,803 'orchestr':681 'output':512,520,566,573,875 'p0/p1/p2/p3':504,537 'pass':348 'path':351,464,786 'pattern':774 'per':106,510,525,580 'persona':155,625 'phase':263 'plan':108,527 'plugin':30,101,119,337,423 'pm':3,10,15,28,34,54,70,150,183,242,288,304,335,342,421,458,484,608,658,679 'pm-artifact-produc':607 'pm-critic':14,69,182,483,657 'pm-release-conductor':303 'pm-skill':27,334,420 'pm-workflow-orchestr':678 'prd':152,267,598,614 'prds':499 'prefer':199 'problem':162 'produc':470,501,610,645 'product':732 'prompt':442,487 'prose':544 'provid':366,787 'qualiti':691 'quality-g':690 're':760 're-run':759 'read':48,429,455,475,770,783,819 'readabl':534,564 'real':590 'recap':159,622 'recent':357,469 'refer':790,851 'referenc':480,821 'references/example.md':583,880 'references/template.md':569,877 'releas':305 'repo':236,247 'repo-level':235 'report':538 'retrospect':168 'return':367 'review':7,88,147,259,276,281,390 'rule':295 'run':5,86,172,194,328,400,595,635,640,761 'runtim':314,870 'section':529,539,556 'see':568,582,703 'semant':203,208 'session':360,473 'set':154,496 'ship':683 'show':588 'skill':29,57,59,137,201,220,243,246,257,264,285,312,324,336,383,422,600,603,611,642,654,668,688,724,728,849 'skill-invoc':200 'skill-utility-pm-critic' 'skill.md':815 'skills/deliver-prd/skill.md':497 'skills/foundation-okr-writer/skill.md':493 'source-product-on-purpose' 'spec':861 'specif':282 'specifi':460 'standard':478,822 'state':248 'statement':163 'status':541,558,578,716 'step':216,316,693 'structur':233,513,524,574 'style':294,816 'sub':18,73,103,113,186,206,223,370,386,425,433,648,661,705,721,855 'sub-ag':17,72,102,112,185,205,222,369,385,424,432,647,660,704,720,854 'subagents/pm-critic.md':49,135,437,858 'suggest':508 'summar':545 'summari':542,725 'support':427,794,812,831 'sweep':300 'synthesi':161,633 'sys':53 'system':441,486 'target':346,457 'templat':876 'time':826 'tool':283 'topic-agent-skills' 'topic-ai-skills' 'topic-claude-code' 'topic-claude-desktop' 'topic-design-sprint' 'topic-foundation-sprint' 'topic-openskills' 'topic-product-management' 'topic-skill-families' 'topic-triple-diamond' 'treat':797 'turn':453,808 'type':491 'unreli':838 'use':143,229,239,260,277,846 'user':79,96,376,463,552,865 'user-fac':864 'util':2,241 'utility-pm-crit':1 'utility-pm-skill-auditor':240 'v2.16.0':731 'v2.17':682 'via':12 'want':145,231,251,274,291,747 'window':396 'windsurf':44,128,409,741 'without':97,418 'work':586,878 'workflow':217,676,680,684 'wrapper':66 'writer':271,618 'yaml':559","prices":[{"id":"4147e37a-0bdf-422e-93f4-196979c52a28","listingId":"b4662567-9b65-4cb5-a500-db752f9e0446","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"product-on-purpose","category":"pm-skills","install_from":"skills.sh"},"createdAt":"2026-05-17T18:54:31.380Z"}],"sources":[{"listingId":"b4662567-9b65-4cb5-a500-db752f9e0446","source":"github","sourceId":"product-on-purpose/pm-skills/utility-pm-critic","sourceUrl":"https://github.com/product-on-purpose/pm-skills/tree/main/skills/utility-pm-critic","isPrimary":false,"firstSeenAt":"2026-05-17T18:54:31.380Z","lastSeenAt":"2026-05-18T18:55:01.699Z"}],"details":{"listingId":"b4662567-9b65-4cb5-a500-db752f9e0446","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"product-on-purpose","slug":"utility-pm-critic","github":{"repo":"product-on-purpose/pm-skills","stars":221,"topics":["agent-skills","ai-skills","claude-code","claude-desktop","design-sprint","foundation-sprint","openskills","product-management","skill-families","triple-diamond"],"license":"apache-2.0","html_url":"https://github.com/product-on-purpose/pm-skills","pushed_at":"2026-05-18T18:13:01Z","description":"55 plug-and-play PM skills for AI agents: 26 Triple Diamond phase + 8 foundation (persona, OKR writer, meeting family, lean canvas) + 6 utility (skill lifecycle) + 15 tool implementing canonical Foundation Sprint + Design Sprint methodologies. Templates, workflows, samples, CI-enforced family contracts. Apache 2.0.","skill_md_sha":"cc49748899922b880f26a1cf993539b183939171","skill_md_path":"skills/utility-pm-critic/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/product-on-purpose/pm-skills/tree/main/skills/utility-pm-critic"},"layout":"multi","source":"github","category":"pm-skills","frontmatter":{"name":"utility-pm-critic","license":"Apache-2.0","description":"Run adversarial review on a PM artifact via the pm-critic sub-agent. Dispatches natively on Claude Code with the pm-skills plugin (invokes @agent-pm-critic); on non-Claude clients (Codex CLI, Cursor, Windsurf, Copilot, Gemini CLI) reads subagents/pm-critic.md and executes the system prompt inline. Returns findings graded P0/P1/P2/P3 with concrete fix suggestions per finding, plus a layered Status Summary section and machine-readable Status YAML block per master plan D26."},"skills_sh_url":"https://skills.sh/product-on-purpose/pm-skills/utility-pm-critic"},"updatedAt":"2026-05-18T18:55:01.699Z"}}