{"id":"9cef7bff-6559-48d7-abe5-1728914b0a97","shortId":"vwKVsu","kind":"skill","title":"dispatching-parallel-agents","tagline":"Dispatches one subagent per independent domain to parallelize investigation/fixes. Use when you have 2+ unrelated failures (e.g., separate failing test files, subsystems, bugs) with no shared state or ordering dependencies.","description":"# Dispatching Parallel Agents\n\nDispatch one agent per independent problem. Let them work concurrently.\n\n## Dispatch Workflow\n\nCopy and track:\n\n```\n- [ ] 1. Identify independent domains\n- [ ] 2. Create focused agent tasks\n- [ ] 3. Dispatch in parallel\n- [ ] 4. Review and integrate\n```\n\n### 1. Identify Independent Domains\n\nGroup failures by what's broken:\n\n- File A tests: Tool approval flow\n- File B tests: Batch completion\n- File C tests: Abort functionality\n\nEach domain is independent—fixing tool approval doesn't affect abort tests.\n\n**Critical check:** If fixing one might fix others → investigate together first (don't parallelize).\n\n### 2. Create Focused Agent Tasks\n\nEach agent needs:\n\n- **Scope:** One test file or subsystem\n- **Goal:** Make these tests pass\n- **Constraints:** Don't change unrelated code\n- **Output:** Summary of findings and fixes\n\n### 3. Dispatch in Parallel\n\nExample (Claude Code):\n\n```typescript\nTask(\"Fix agent-tool-abort.test.ts failures\")\nTask(\"Fix batch-completion-behavior.test.ts failures\")\nTask(\"Fix tool-approval-race-conditions.test.ts failures\")\n```\n\n### 4. Review and Integrate\n\n1. Read each agent's summary\n2. Check for conflicts (same files edited?)\n   - If two agents touched the same file → stop and re-scope (one owner per file)\n3. Run full test suite\n4. If failures:\n   - Check for merge conflicts → resolve manually\n   - If no conflicts → investigate as new failures\n5. Repeat until green\n\n## Agent Prompt Template\n\n```markdown\nFix the [N] failing tests in [file path]:\n\n1. \"[test name]\" - [error summary]\n2. \"[test name]\" - [error summary]\n\nContext: [relevant background, e.g., \"These are timing/race condition issues\"]\n\nYour task:\n1. Read the test file, understand what each test verifies\n2. Identify root cause—timing issues or actual bugs?\n3. Fix by [preferred approach, e.g., \"replacing arbitrary timeouts with event-based waiting\"]\n\nDo NOT: [anti-patterns, e.g., \"just increase timeouts—find the real issue\"]\n\nReturn: Summary of root cause and changes made.\n```\n\n## Common Mistakes\n\n| ❌ Bad | ✅ Good |\n|--------|---------|\n| \"Fix all the tests\" | \"Fix agent-tool-abort.test.ts\" |\n| \"Fix the race condition\" | Paste error messages + test names |\n| No constraints | \"Do NOT change production code\" |\n| \"Fix it\" | \"Return summary of root cause and changes\" |\n\n## Example\n\n**Scenario:** 6 test failures across 3 files after major refactoring.\n\n**Failures:**\n\n- agent-tool-abort.test.ts: 3 failures (timing issues)\n- batch-completion-behavior.test.ts: 2 failures (tools not executing)\n- tool-approval-race-conditions.test.ts: 1 failure (execution count = 0)\n\n**Decision:** Independent domains—abort logic separate from batch completion separate from race conditions.\n\n**Dispatch:**\n\n```\nAgent 1 → Fix agent-tool-abort.test.ts\nAgent 2 → Fix batch-completion-behavior.test.ts\nAgent 3 → Fix tool-approval-race-conditions.test.ts\n```\n\n**Results:**\n\n- Agent 1: Replaced timeouts with event-based waiting\n- Agent 2: Fixed event structure bug (threadId in wrong place)\n- Agent 3: Added wait for async tool execution\n\n**Integration:** All fixes independent, no conflicts, full suite green.","tags":["dispatching","parallel","agents","agent","skills","library","codingcossack","agent-framework","agent-skills","agent-system","agent-workflow","agentic-workflow"],"capabilities":["skill","source-codingcossack","skill-dispatching-parallel-agents","topic-agent-framework","topic-agent-skills","topic-agent-system","topic-agent-workflow","topic-agentic-workflow","topic-ai-agents","topic-anthropic","topic-claude","topic-claude-code","topic-claude-skills","topic-claude-skills-hub","topic-claude-skills-libary"],"categories":["agent-skills-library"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/CodingCossack/agent-skills-library/dispatching-parallel-agents","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add CodingCossack/agent-skills-library","source_repo":"https://github.com/CodingCossack/agent-skills-library","install_from":"skills.sh"}},"qualityScore":"0.458","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 17 github stars · SKILL.md body (3,132 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":"2026-04-23T07:01:19.073Z","embedding":null,"createdAt":"2026-04-18T23:06:35.153Z","updatedAt":"2026-04-23T07:01:19.073Z","lastSeenAt":"2026-04-23T07:01:19.073Z","tsv":"'0':381 '1':53,70,177,243,264,377,397,410 '2':18,57,122,183,248,274,371,401,419 '3':62,153,206,283,359,366,405,429 '4':66,173,211 '5':227 '6':355 'abort':94,106,385 'across':358 'actual':281 'ad':430 'affect':105 'agent':4,37,40,60,125,128,180,192,231,396,400,404,409,418,428 'agent-tool-abort.test.ts':163,327,365,399 'anti':300 'anti-pattern':299 'approach':287 'approv':84,102 'arbitrari':290 'async':433 'b':87 'background':255 'bad':320 'base':295,416 'batch':89,389 'batch-completion-behavior.test.ts':167,370,403 'broken':79 'bug':27,282,423 'c':92 'caus':277,314,350 'chang':144,316,341,352 'check':109,184,214 'claud':158 'code':146,159,343 'common':318 'complet':90,390 'concurr':47 'condit':260,331,394 'conflict':186,217,222,441 'constraint':141,338 'context':253 'copi':50 'count':380 'creat':58,123 'critic':108 'decis':382 'depend':34 'dispatch':2,5,35,38,48,63,154,395 'dispatching-parallel-ag':1 'doesn':103 'domain':10,56,73,97,384 'e.g':21,256,288,302 'edit':189 'error':246,251,333 'event':294,415,421 'event-bas':293,414 'exampl':157,353 'execut':375,379,435 'fail':23,238 'failur':20,75,164,168,172,213,226,357,364,367,372,378 'file':25,80,86,91,133,188,196,205,241,268,360 'find':150,306 'first':118 'fix':100,111,114,152,162,166,170,235,284,322,326,328,344,398,402,406,420,438 'flow':85 'focus':59,124 'full':208,442 'function':95 'goal':136 'good':321 'green':230,444 'group':74 'identifi':54,71,275 'increas':304 'independ':9,42,55,72,99,383,439 'integr':69,176,436 'investig':116,223 'investigation/fixes':13 'issu':261,279,309,369 'let':44 'logic':386 'made':317 'major':362 'make':137 'manual':219 'markdown':234 'merg':216 'messag':334 'might':113 'mistak':319 'n':237 'name':245,250,336 'need':129 'new':225 'one':6,39,112,131,202 'order':33 'other':115 'output':147 'owner':203 'parallel':3,12,36,65,121,156 'pass':140 'past':332 'path':242 'pattern':301 'per':8,41,204 'place':427 'prefer':286 'problem':43 'product':342 'prompt':232 'race':330,393 're':200 're-scop':199 'read':178,265 'real':308 'refactor':363 'relev':254 'repeat':228 'replac':289,411 'resolv':218 'result':408 'return':310,346 'review':67,174 'root':276,313,349 'run':207 'scenario':354 'scope':130,201 'separ':22,387,391 'share':30 'skill' 'skill-dispatching-parallel-agents' 'source-codingcossack' 'state':31 'stop':197 'structur':422 'subag':7 'subsystem':26,135 'suit':210,443 'summari':148,182,247,252,311,347 'task':61,126,161,165,169,263 'templat':233 'test':24,82,88,93,107,132,139,209,239,244,249,267,272,325,335,356 'threadid':424 'time':278,368 'timeout':291,305,412 'timing/race':259 'togeth':117 'tool':83,101,373,434 'tool-approval-race-conditions.test.ts':171,376,407 'topic-agent-framework' 'topic-agent-skills' 'topic-agent-system' 'topic-agent-workflow' 'topic-agentic-workflow' 'topic-ai-agents' 'topic-anthropic' 'topic-claude' 'topic-claude-code' 'topic-claude-skills' 'topic-claude-skills-hub' 'topic-claude-skills-libary' 'touch':193 'track':52 'two':191 'typescript':160 'understand':269 'unrel':19,145 'use':14 'verifi':273 'wait':296,417,431 'work':46 'workflow':49 'wrong':426","prices":[{"id":"9d3ef5c6-e37b-4783-b39a-d7b448d25831","listingId":"9cef7bff-6559-48d7-abe5-1728914b0a97","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"CodingCossack","category":"agent-skills-library","install_from":"skills.sh"},"createdAt":"2026-04-18T23:06:35.153Z"}],"sources":[{"listingId":"9cef7bff-6559-48d7-abe5-1728914b0a97","source":"github","sourceId":"CodingCossack/agent-skills-library/dispatching-parallel-agents","sourceUrl":"https://github.com/CodingCossack/agent-skills-library/tree/main/skills/dispatching-parallel-agents","isPrimary":false,"firstSeenAt":"2026-04-18T23:06:35.153Z","lastSeenAt":"2026-04-23T07:01:19.073Z"}],"details":{"listingId":"9cef7bff-6559-48d7-abe5-1728914b0a97","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"CodingCossack","slug":"dispatching-parallel-agents","github":{"repo":"CodingCossack/agent-skills-library","stars":17,"topics":["agent-framework","agent-skills","agent-system","agent-workflow","agentic-workflow","ai-agents","anthropic","claude","claude-code","claude-skills","claude-skills-hub","claude-skills-libary","code-review","codex","context-engineering","debugging","developer-workflow"],"license":null,"html_url":"https://github.com/CodingCossack/agent-skills-library","pushed_at":"2026-01-03T20:02:38Z","description":"Coding agent skills library for programming workflows | Claude Skills, Codex Skills | Forked from obra/superpower","skill_md_sha":"562c5b830cb6180afb10cd75f712e1ba4203b27b","skill_md_path":"skills/dispatching-parallel-agents/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/CodingCossack/agent-skills-library/tree/main/skills/dispatching-parallel-agents"},"layout":"multi","source":"github","category":"agent-skills-library","frontmatter":{"name":"dispatching-parallel-agents","description":"Dispatches one subagent per independent domain to parallelize investigation/fixes. Use when you have 2+ unrelated failures (e.g., separate failing test files, subsystems, bugs) with no shared state or ordering dependencies."},"skills_sh_url":"https://skills.sh/CodingCossack/agent-skills-library/dispatching-parallel-agents"},"updatedAt":"2026-04-23T07:01:19.073Z"}}