[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AGXM164VCP0ANC7FQXVHT2":3,"skill-same:01M7AGXM164VCP0ANC7FQXVHT2":59},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M7AGXM164VCP0ANC7FQXVHT2","loop-engineering",null,"Execute non-trivial software work through small, observable build-test-observe-adjust loops with explicit hypotheses, evidence, outcomes, and next actions. Use throughout feature implementation, debugging, refactoring, migrations, performance work, and complex verification where feedback must guide each subsequent step. Use when work is stalling, changes are growing too large, failures repeat, or an engineering ledger is required.","projectamazonph\u002Fagent-skill-framework","projectamazonph","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F286085559?v=4","01M79QFF84668ZQCDVGZXP34RF","Ryan Roland Dabao ","01M7AGNPF0TY0MJR8KX7BGQK98",0,"MIT","https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Floop-engineering\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: loop-engineering\ndescription: Execute non-trivial software work through small, observable build-test-observe-adjust loops with explicit hypotheses, evidence, outcomes, and next actions. Use throughout feature implementation, debugging, refactoring, migrations, performance work, and complex verification where feedback must guide each subsequent step. Use when work is stalling, changes are growing too large, failures repeat, or an engineering ledger is required.\n---\n\n# Loop Engineering\n\n## Mission\n\nOwn execution cadence and learning. Keep each loop small, observable, reversible with precise edits, and driven by evidence.\n\nDo not own detailed test method, architecture policy, or final approval. Route behavior changes to `test-driven-development`, structural decisions to `solid-design`, and completion to `quality-gates` through the orchestrator.\n\n## Suite Protocol\n\nWhen another skill routes work here, read [references\u002Fsuite-contract.md](references\u002Fsuite-contract.md). Accept the Task Packet and return the standard Handoff Packet. Each loop adds evidence and a next action; it does not independently declare the overall task complete.\n\n## Open a Loop\n\nDefine:\n\n```text\nLoop: \u003Cnumber or short ID>\nHypothesis: \u003Cwhat this step should prove or change>\nScope: \u003Cone behavior, diagnostic, or structural move>\nEvidence plan: \u003Ctest, inspection, metric, build, or runtime signal>\nStop condition: \u003Cobservable success, failure, or split threshold>\n```\n\nReject a loop that contains several independently verifiable outcomes. Split by behavior, dependency, caller group, migration batch, or risk.\n\n## Run the Cycle\n\n### 1. Observe\n\nInspect current code, tests, logs, runtime behavior, metrics, or errors. Separate known baseline failures from new failures. Never begin from an assumed state.\n\n### 2. Frame\n\nChoose the smallest uncertainty or behavior that blocks progress. State a falsifiable hypothesis and the evidence needed to decide.\n\n### 3. Execute\n\n- For new behavior or a defect, invoke `test-driven-development` and complete one valid Red–Green–Refactor cycle.\n- For a structural move, obtain a boundary decision from `solid-design`, preserve behavior with tests, and move one responsibility, dependency, contract, or caller group.\n- For diagnosis, change one variable or add one observation point. Do not mix diagnosis with a broad fix.\n- For bulk work, run one representative or low-risk batch before scaling.\n\n### 4. Measure\n\nRun the focused evidence check, then proportionate broader checks. Capture actual output, not memory or expectation.\n\n### 5. Compare\n\nClassify the loop:\n\n- **Confirmed:** Evidence matches the hypothesis and acceptance behavior.\n- **Refuted:** Evidence disproves the hypothesis.\n- **Inconclusive:** The observation cannot distinguish causes.\n- **Blocked:** A missing dependency, permission, decision, or broken baseline prevents a valid result.\n\n### 6. Adjust\n\nChoose exactly one next action:\n\n- Advance to the next behavior or structural move.\n- Tighten the test or observation.\n- Reduce the scope.\n- Return to architecture shaping.\n- Repair the baseline or environment.\n- Escalate a material user decision.\n- Stop because acceptance is satisfied or the work is blocked.\n\nDo not stack speculative fixes after a refuted or inconclusive loop.\n\n## Keep the Ledger\n\nUse an existing build log, engineering diary, task record, or project journal. If none exists, retain a concise ledger for the final report:\n\n| Loop | Hypothesis | Action | Evidence | Outcome | Next |\n| --- | --- | --- | --- | --- | --- |\n\nRecord exact commands and meaningful results. Do not dump noisy logs when a precise summary and failure excerpt are sufficient.\n\nFor long-running work, send the user brief progress updates at meaningful loop boundaries without turning the ledger into play-by-play commentary.\n\n## Control Loop Size\n\nSplit the loop when:\n\n- More than one acceptance behavior is changing.\n- Several production modules must change for unrelated reasons.\n- A test failure has more than one plausible cause.\n- Review becomes difficult because the diff mixes behavior and structure.\n- A migration batch cannot be independently verified or rolled back.\n- The success condition cannot be stated in one sentence.\n\nOne loop may touch multiple files when they form one atomic behavior or contract change. File count alone does not define loop size.\n\n## Recover from Failure\n\nRead [references\u002Floop-recovery.md](references\u002Floop-recovery.md) when loops repeat, focused checks pass but broader checks fail, the environment is unreliable, or the next action is unclear.\n\nUse precise patches to remove only the failed experiment's changes. Preserve unrelated user work. Never use destructive resets as a loop shortcut.\n\n## Stop Correctly\n\nStop iteration when one condition holds:\n\n- Requested acceptance behavior is satisfied and evidence is ready for `quality-gates`.\n- A material decision or new authority is required from the user.\n- The environment cannot produce trustworthy evidence.\n- Continuing would expand scope beyond the request.\n\nDo not stop merely because the first plausible implementation compiles or one focused test passes.\n\n## Return the Handoff\n\nReturn to the orchestrator:\n\n```text\nCompleted loops: \u003CIDs and outcomes>\nAcceptance status: \u003Csatisfied | partial | blocked>\nFocused evidence: \u003Cresults>\nBroader observation: \u003Cresults>\nArchitecture questions: \u003Cnone or route to SOLID>\nResidual risk: \u003Cspecific gap>\nRecommended next phase: \u003Cshape | red | observe | gate | blocked>\n```\n","# Loop Engineering\n\n## Mission\n\nOwn execution cadence and learning. Keep each loop small, observable, reversible with precise edits, and driven by evidence.\n\nDo not own detailed test method, architecture policy, or final approval. Route behavior changes to `test-driven-development`, structural decisions to `solid-design`, and completion to `quality-gates` through the orchestrator.\n\n## Suite Protocol\n\nWhen another skill routes work here, read [references\u002Fsuite-contract.md](references\u002Fsuite-contract.md). Accept the Task Packet and return the standard Handoff Packet. Each loop adds evidence and a next action; it does not independently declare the overall task complete.\n\n## Open a Loop\n\nDefine:\n\n```text\nLoop: \u003Cnumber or short ID>\nHypothesis: \u003Cwhat this step should prove or change>\nScope: \u003Cone behavior, diagnostic, or structural move>\nEvidence plan: \u003Ctest, inspection, metric, build, or runtime signal>\nStop condition: \u003Cobservable success, failure, or split threshold>\n```\n\nReject a loop that contains several independently verifiable outcomes. Split by behavior, dependency, caller group, migration batch, or risk.\n\n## Run the Cycle\n\n### 1. Observe\n\nInspect current code, tests, logs, runtime behavior, metrics, or errors. Separate known baseline failures from new failures. Never begin from an assumed state.\n\n### 2. Frame\n\nChoose the smallest uncertainty or behavior that blocks progress. State a falsifiable hypothesis and the evidence needed to decide.\n\n### 3. Execute\n\n- For new behavior or a defect, invoke `test-driven-development` and complete one valid Red–Green–Refactor cycle.\n- For a structural move, obtain a boundary decision from `solid-design`, preserve behavior with tests, and move one responsibility, dependency, contract, or caller group.\n- For diagnosis, change one variable or add one observation point. Do not mix diagnosis with a broad fix.\n- For bulk work, run one representative or low-risk batch before scaling.\n\n### 4. Measure\n\nRun the focused evidence check, then proportionate broader checks. Capture actual output, not memory or expectation.\n\n### 5. Compare\n\nClassify the loop:\n\n- **Confirmed:** Evidence matches the hypothesis and acceptance behavior.\n- **Refuted:** Evidence disproves the hypothesis.\n- **Inconclusive:** The observation cannot distinguish causes.\n- **Blocked:** A missing dependency, permission, decision, or broken baseline prevents a valid result.\n\n### 6. Adjust\n\nChoose exactly one next action:\n\n- Advance to the next behavior or structural move.\n- Tighten the test or observation.\n- Reduce the scope.\n- Return to architecture shaping.\n- Repair the baseline or environment.\n- Escalate a material user decision.\n- Stop because acceptance is satisfied or the work is blocked.\n\nDo not stack speculative fixes after a refuted or inconclusive loop.\n\n## Keep the Ledger\n\nUse an existing build log, engineering diary, task record, or project journal. If none exists, retain a concise ledger for the final report:\n\n| Loop | Hypothesis | Action | Evidence | Outcome | Next |\n| --- | --- | --- | --- | --- | --- |\n\nRecord exact commands and meaningful results. Do not dump noisy logs when a precise summary and failure excerpt are sufficient.\n\nFor long-running work, send the user brief progress updates at meaningful loop boundaries without turning the ledger into play-by-play commentary.\n\n## Control Loop Size\n\nSplit the loop when:\n\n- More than one acceptance behavior is changing.\n- Several production modules must change for unrelated reasons.\n- A test failure has more than one plausible cause.\n- Review becomes difficult because the diff mixes behavior and structure.\n- A migration batch cannot be independently verified or rolled back.\n- The success condition cannot be stated in one sentence.\n\nOne loop may touch multiple files when they form one atomic behavior or contract change. File count alone does not define loop size.\n\n## Recover from Failure\n\nRead [references\u002Floop-recovery.md](references\u002Floop-recovery.md) when loops repeat, focused checks pass but broader checks fail, the environment is unreliable, or the next action is unclear.\n\nUse precise patches to remove only the failed experiment's changes. Preserve unrelated user work. Never use destructive resets as a loop shortcut.\n\n## Stop Correctly\n\nStop iteration when one condition holds:\n\n- Requested acceptance behavior is satisfied and evidence is ready for `quality-gates`.\n- A material decision or new authority is required from the user.\n- The environment cannot produce trustworthy evidence.\n- Continuing would expand scope beyond the request.\n\nDo not stop merely because the first plausible implementation compiles or one focused test passes.\n\n## Return the Handoff\n\nReturn to the orchestrator:\n\n```text\nCompleted loops: \u003CIDs and outcomes>\nAcceptance status: \u003Csatisfied | partial | blocked>\nFocused evidence: \u003Cresults>\nBroader observation: \u003Cresults>\nArchitecture questions: \u003Cnone or route to SOLID>\nResidual risk: \u003Cspecific gap>\nRecommended next phase: \u003Cshape | red | observe | gate | blocked>\n```\n",{"name":5,"description":7},"main","fbd23e0a0b843f9cc3f1f69d5317659ef0dfad0d","skills\u002Fsuite-core\u002Floop-engineering","https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework","https:\u002F\u002Fgithub.com\u002Fprojectamazonph","Amazon PPC specialist focused on campaign systems, automation, VA training, and practical eCommerce tools.","2026-09-22 23:51:13",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add projectamazonph\u002Fagent-skill-framework --skill loop-engineering",{"tool":37,"command":38},"bunx","bunx skills add projectamazonph\u002Fagent-skill-framework --skill loop-engineering",{"tool":40,"command":41},"pnpm","pnpm dlx skills add projectamazonph\u002Fagent-skill-framework --skill loop-engineering","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Floop-engineering\u002FSKILL.md",true,[45,51,55],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"agents\u002Fopenai.yaml","REFERENCE","参考文档",328,"https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Floop-engineering\u002Fagents\u002Fopenai.yaml",{"path":52,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Floop-recovery.md",2345,"https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Floop-engineering\u002Freferences\u002Floop-recovery.md",{"path":56,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Fsuite-contract.md",2179,"https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Floop-engineering\u002Freferences\u002Fsuite-contract.md",[60,67,74,81,88,95],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M7AGYHHPDMHED8S7AVBN4JJ7","test-driven-development","Drive one observable software behavior at a time through a valid Test-Driven Development cycle: select the test, prove Red for the intended reason, implement the smallest Green change, and preserve the test as regression evidence. Use before adding production behavior, fixing defects, changing contracts, or safely characterizing legacy behavior. Also use when tests are brittle, over-mocked, skipped, weakened, or passing without proving the requirement.","https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M7AGYB9EV23KZN66XG70RXRP","suite-orchestrator","Route complex work across the complete Agent Skill Suite while enforcing content preservation, explicit scope, correct specialist timing, evidence-backed completion, and durable state. Use when a task involves coding, agent operations, repository bootstrapping, conversation compilation, artifact generation, expert review, persistent memory, or several of these together. Also use when the correct skill is unclear or work must pass through multiple specialist skills in a controlled sequence.","https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Fsuite-orchestrator\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7AGXX0ZJR3K8QNV7J38NF3X","solid-design","Design, refactor, and review object-oriented or modular code with the SOLID principles while preserving behavior and avoiding needless abstraction. Use when defining responsibilities, extension seams, subtype contracts, consumer interfaces, dependency direction, modules, services, strategies, plugins, adapters, repositories, or architecture boundaries; also use when test friction, tangled dependencies, large classes, unsupported methods, framework leakage, or repeated variant branches signal a structural problem.","https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Fsolid-design\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M7AGXRGA3HE2EFAEN81YP7BQ","quality-gates","Verify engineering work against required behavioral, architectural, regression, build, security, compatibility, documentation, and diff-hygiene evidence before completion. Use after a code change, defect fix, refactor, migration, generated-code update, or implementation plan appears complete; also use for read-only code reviews and release-readiness checks. Return PASS or BLOCKED without implementing the remediation.","https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Fquality-gates\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":19},"01M7AGXFMA0Q813X6186JXKV7H","local-memory","Maintain fast, durable, curated local memory for AI agents on Android, Termux, Linux, macOS, Windows, or other lightweight environments. Use when an agent must remember stable preferences, project facts, decisions, procedures, failure resolutions, and task state across sessions; retrieve relevant memories before planning; resolve conflicting or stale memories; or provide explicit remember, search, pin, update, forget, status, and export controls without a cloud dependency.","https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Flocal-memory\u002FSKILL.md",[],[],{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":19},"01M7AGXB090RWQ27P5BH4HPGN4","implementation-council","Pressure-test codebases, specifications, architecture notes, implementation plans, test strategies, migrations, and deployment plans through a structured council of fixed engineering leaders and task-specific specialists. Use when the user requests an expert panel, adversarial review, debate, gap analysis, implementation improvement, production-readiness assessment, or deployment verdict. Require evidence, disagreement, rebuttal, prioritized gaps, and a clear readiness decision.","https:\u002F\u002Fgithub.com\u002Fprojectamazonph\u002Fagent-skill-framework\u002Fblob\u002Fmain\u002Fskills\u002Fsuite-core\u002Fimplementation-council\u002FSKILL.md",[],[]]