[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M698X5677QF1CY4Y2WT6N38N":3,"skill-same:01M698X5677QF1CY4Y2WT6N38N":73},{"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":28,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":57,"updatedAt":58,"install":59,"resources":72},"01M698X5677QF1CY4Y2WT6N38N","code-review",null,"Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.","google\u002Ffilament","google","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1342004?v=4","01M698VYEFDM0VJBZDYYJ2T83S","Google","01M698W8SGYAMBEGYBK16J57R1",20747,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament\u002Fblob\u002Fmain\u002Fskills\u002Fcode_review\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000007","代码审查","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"code review","review","code","diffs","memory","architecture","protocol","concurrency","review code","model","use","branches","systems","prs","adversarial","2026-11-10 10:12:30",false,"---\nname: code-review\ndescription: >\n  Adversarial systems, concurrency, memory-model, and architecture code review protocol.\n  Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.\n---\n\n# Systems & Concurrency Code Review Protocol\n\nThis skill defines the mandatory protocol for conducting rigorous, adversarial code reviews on Filament source code. Standard single-pass reviews focusing on style or superficial idioms are strictly prohibited. Reviews must prioritize **correctness, concurrency invariants, memory models, lifecycle safety, API contracts, arithmetic boundaries, and test integrity**.\n\n---\n\n## 1. Core Review Philosophy\n\n1. **Adversarial Stance**: Assume the code contains subtle race conditions, memory ordering errors, and unhandled edge cases until proven otherwise.\n2. **Zero Superficial Feedback**: Do not waste review bandwidth on formatting, whitespace, or obvious variable naming unless it actively impairs correctness or violates [CODE_STYLE.md](..\u002F..\u002FCODE_STYLE.md).\n3. **Cross-Function & State-Machine Tracing**: Never review functions in isolation. Concurrency bugs (e.g., lost wakeups, ABA races) live at the intersection of multiple functions (e.g., `put()`, `wakeOne()`, `waitForWork()`, `loop()`). Trace full execution state machines across caller\u002Fcallee boundaries.\n4. **False-Positive Elimination**: Before reporting a finding, attempt to disprove it by searching for compensating synchronization, refcount credits, or structural invariants elsewhere in the file or call chain.\n\n---\n\n## 2. Mandatory Analytical Lenses (The 6 Pillars)\n\nEvery review must systematically evaluate the change through these six analytical lenses:\n\n### Pillar 1: Concurrency, Synchronization & Memory Models\n*   **Atomic Memory Ordering & Barrier Direction**:\n    *   Inspect every `std::atomic` operation and its `std::memory_order` (`relaxed`, `acquire`, `release`, `acq_rel`, `seq_cst`).\n    *   Verify barrier directions: `memory_order_release` prevents *prior* memory writes from sinking below the release store, but does **not** prevent *subsequent* non-atomic writes from floating above it. `memory_order_acquire` prevents *subsequent* reads from floating above the acquire load, but does **not** protect prior reads.\n*   **Lost Wakeups & Condition Variable Invariants**:\n    *   Check if condition variable signals (`notify_one()` \u002F `wakeOne()`) or waiting loops (`wait()`) are gated by unlocked relaxed loads (e.g., relaxed check of waiter counts).\n    *   Verify that the happens-before relationship between producer publication and consumer sleep is never bypassed.\n*   **ABA, Slot-Reuse & Torn Reads**:\n    *   In wait-free, lock-free, or seqlock data structures, trace multi-step read-then-validate patterns.\n    *   Ensure that slot identity (e.g., thread ID, generation counter) is re-validated *after* copying payload data, not just the slot state, to prevent returning payload data read during an erase\u002Fre-emplace.\n*   **Data Races in Lock-Free \u002F Wait-Free APIs**:\n    *   Verify that plain (non-atomic) fields are never read concurrently with plain writes in other threads without a formal synchronization barrier.\n*   **Deadlock & Lock Invariants**:\n    *   Verify lock acquisition order. Ensure code uses `utils::Mutex` \u002F `utils::Condition` with `LockGuard const` as mandated by [skills\u002Fcpp_static_thread_safety](..\u002Fcpp_static_thread_safety\u002FSKILL.md).\n\n### Pillar 2: Arithmetic, Narrowing & Numerical Boundaries\n*   **Integer Truncation & Narrowing Conversions**:\n    *   Check all numeric assignments and casts between types of differing bit widths (e.g., `uint32_t` $\\to$ `uint16_t`, `size_t` $\\to$ `uint32_t`).\n    *   Determine what happens when values equal or exceed the narrower type's capacity (e.g., exact multiples of 65,536 truncating to 0 and inducing infinite loops).\n*   **Boundary Conditions**:\n    *   Evaluate behavior at `count == 0`, `count == 1`, $2^{16}$, $2^{32}$, `INT_MAX`, and overflow\u002Funderflow points.\n*   **Resource & Pool Budgets**:\n    *   Ensure budget calculations cannot silently drop minimum guarantees (e.g., combined thread limits reducing worker pool size to 0 without assertion or fallback).\n\n### Pillar 3: API Contracts & Behavioral Deltas\n*   **Semantic Regressions**:\n    *   Compare old vs. new contracts. Did zero-sized inputs (`count == 0`), empty containers, or null arguments change behavior (e.g., skipping a functor that previously ran once)?\n*   **Trait Specializations & SFINAE \u002F Concepts**:\n    *   Verify that template traits (e.g., `SplitterTraits`) do not silently ignore legacy interfaces or custom types without a compilation error or fallback.\n*   **Doc-Comment vs. Reality**:\n    *   Verify that documented thread-safety guarantees (e.g., \"wait-free\", \"thread-safe for concurrent readers\") are actually fulfilled by the implementation.\n\n### Pillar 4: Object Lifetime, Teardown & RAII\n*   **Thread Lifecycle & Cleanup**:\n    *   Verify that adopted threads or registered slots properly decrement active counters upon exit\u002Femancipation (`emancipate()`), preventing capacity degradation.\n*   **Asynchronous Lambda Captures**:\n    *   Ensure captured references (`[&]`) or raw pointers (`this`) outlive background job execution.\n*   **Refcount Draining & Destruction Races**:\n    *   Verify that root jobs \u002F parent structures hold an explicit refcount credit during worker dispatch so refcounts cannot hit zero mid-execution on worker failure.\n\n### Pillar 5: Test Integrity & False-Pass Audits\n*   **Tautological Assertions**:\n    *   Inspect test assertions to ensure they are not tautologies (e.g., assertions that can never evaluate to false).\n*   **Concurrency Stress & Race Coverage**:\n    *   Verify that multi-threaded tests genuinely create contention on the guarded path rather than executing deterministically or passing due to timing coincidence.\n*   **Build Wiring**:\n    *   Verify new test files are wired into the corresponding `CMakeLists.txt`.\n\n### Pillar 6: Architectural Completeness & Scaling\n*   **Feature Completeness**:\n    *   When introducing a new capability (e.g., adopted worker threads), verify that all sizing, partitioning, and helper calculations (e.g., `parallel_for` helper job counts) scale with the new capability rather than only querying legacy state.\n\n---\n\n## 3. Standard Review Output Format\n\nReviews must be structured using the following format:\n\n```markdown\n\u003CFile Path>\n  - \u003Cline> [\u003Ccategory>] \u003CConcise finding description detailing the exact failure mechanism.>\n\n...\n\nReview Summary:\n- **The Blocker**: Identify the single most critical blocking issue, explaining the thread interleaving or failure sequence.\n- **Intent vs. Code**: Assess whether the PR achieves its architectural goals and note any systematic omissions.\n- **Explicitly Checked & Ruled Out**: List non-obvious complex areas investigated and verified as safe\u002Fcorrect (proves thorough inspection and eliminates false positives).\n- **Minor Cleanups**: Note dead code, minor un-decremented tracking counters, or triplicated logic below the blocking finding bar.\n```\n\n### Category Tags:\n*   `[correctness]` — Logic errors, data races, memory ordering bugs, lost wakeups, narrowing bugs, crashes.\n*   `[efficiency]` — Scaling bottlenecks, false sharing, unnecessary allocations, missed parallelism.\n*   `[api-contract]` — Broken traits, dropped interface support, violated preconditions\u002Fpostconditions.\n*   `[behavior-change]` — Silent changes to edge-case behavior (e.g., `count == 0`).\n*   `[test-coverage]` — Tautological tests, missing failure checks, unwired tests.\n\n---\n\n## 4. Review Execution Workflow for AI Agents\n\nWhen instructed to review code, a branch, a commit range, or a PR:\n\n1. **Discover Changes**:\n   ```bash\n   git diff \u003Cbase_branch>...\u003Ctarget_branch>\n   # or\n   git log -n \u003Ccount> --stat\n   ```\n2. **Read Related Context**:\n   Read the modified files alongside their callers and dependencies using `view_file` to understand the complete cross-function state machine.\n3. **Execute the 6 Analytical Pillars**:\n   Systematically audit the diff against each pillar.\n4. **Draft & Filter Findings**:\n   Verify every potential finding against the codebase to eliminate false positives.\n5. **Format Report**:\n   Output the structured findings following the format in Section 3.\n","# Systems & Concurrency Code Review Protocol\n\nThis skill defines the mandatory protocol for conducting rigorous, adversarial code reviews on Filament source code. Standard single-pass reviews focusing on style or superficial idioms are strictly prohibited. Reviews must prioritize **correctness, concurrency invariants, memory models, lifecycle safety, API contracts, arithmetic boundaries, and test integrity**.\n\n---\n\n## 1. Core Review Philosophy\n\n1. **Adversarial Stance**: Assume the code contains subtle race conditions, memory ordering errors, and unhandled edge cases until proven otherwise.\n2. **Zero Superficial Feedback**: Do not waste review bandwidth on formatting, whitespace, or obvious variable naming unless it actively impairs correctness or violates [CODE_STYLE.md](..\u002F..\u002FCODE_STYLE.md).\n3. **Cross-Function & State-Machine Tracing**: Never review functions in isolation. Concurrency bugs (e.g., lost wakeups, ABA races) live at the intersection of multiple functions (e.g., `put()`, `wakeOne()`, `waitForWork()`, `loop()`). Trace full execution state machines across caller\u002Fcallee boundaries.\n4. **False-Positive Elimination**: Before reporting a finding, attempt to disprove it by searching for compensating synchronization, refcount credits, or structural invariants elsewhere in the file or call chain.\n\n---\n\n## 2. Mandatory Analytical Lenses (The 6 Pillars)\n\nEvery review must systematically evaluate the change through these six analytical lenses:\n\n### Pillar 1: Concurrency, Synchronization & Memory Models\n*   **Atomic Memory Ordering & Barrier Direction**:\n    *   Inspect every `std::atomic` operation and its `std::memory_order` (`relaxed`, `acquire`, `release`, `acq_rel`, `seq_cst`).\n    *   Verify barrier directions: `memory_order_release` prevents *prior* memory writes from sinking below the release store, but does **not** prevent *subsequent* non-atomic writes from floating above it. `memory_order_acquire` prevents *subsequent* reads from floating above the acquire load, but does **not** protect prior reads.\n*   **Lost Wakeups & Condition Variable Invariants**:\n    *   Check if condition variable signals (`notify_one()` \u002F `wakeOne()`) or waiting loops (`wait()`) are gated by unlocked relaxed loads (e.g., relaxed check of waiter counts).\n    *   Verify that the happens-before relationship between producer publication and consumer sleep is never bypassed.\n*   **ABA, Slot-Reuse & Torn Reads**:\n    *   In wait-free, lock-free, or seqlock data structures, trace multi-step read-then-validate patterns.\n    *   Ensure that slot identity (e.g., thread ID, generation counter) is re-validated *after* copying payload data, not just the slot state, to prevent returning payload data read during an erase\u002Fre-emplace.\n*   **Data Races in Lock-Free \u002F Wait-Free APIs**:\n    *   Verify that plain (non-atomic) fields are never read concurrently with plain writes in other threads without a formal synchronization barrier.\n*   **Deadlock & Lock Invariants**:\n    *   Verify lock acquisition order. Ensure code uses `utils::Mutex` \u002F `utils::Condition` with `LockGuard const` as mandated by [skills\u002Fcpp_static_thread_safety](..\u002Fcpp_static_thread_safety\u002FSKILL.md).\n\n### Pillar 2: Arithmetic, Narrowing & Numerical Boundaries\n*   **Integer Truncation & Narrowing Conversions**:\n    *   Check all numeric assignments and casts between types of differing bit widths (e.g., `uint32_t` $\\to$ `uint16_t`, `size_t` $\\to$ `uint32_t`).\n    *   Determine what happens when values equal or exceed the narrower type's capacity (e.g., exact multiples of 65,536 truncating to 0 and inducing infinite loops).\n*   **Boundary Conditions**:\n    *   Evaluate behavior at `count == 0`, `count == 1`, $2^{16}$, $2^{32}$, `INT_MAX`, and overflow\u002Funderflow points.\n*   **Resource & Pool Budgets**:\n    *   Ensure budget calculations cannot silently drop minimum guarantees (e.g., combined thread limits reducing worker pool size to 0 without assertion or fallback).\n\n### Pillar 3: API Contracts & Behavioral Deltas\n*   **Semantic Regressions**:\n    *   Compare old vs. new contracts. Did zero-sized inputs (`count == 0`), empty containers, or null arguments change behavior (e.g., skipping a functor that previously ran once)?\n*   **Trait Specializations & SFINAE \u002F Concepts**:\n    *   Verify that template traits (e.g., `SplitterTraits`) do not silently ignore legacy interfaces or custom types without a compilation error or fallback.\n*   **Doc-Comment vs. Reality**:\n    *   Verify that documented thread-safety guarantees (e.g., \"wait-free\", \"thread-safe for concurrent readers\") are actually fulfilled by the implementation.\n\n### Pillar 4: Object Lifetime, Teardown & RAII\n*   **Thread Lifecycle & Cleanup**:\n    *   Verify that adopted threads or registered slots properly decrement active counters upon exit\u002Femancipation (`emancipate()`), preventing capacity degradation.\n*   **Asynchronous Lambda Captures**:\n    *   Ensure captured references (`[&]`) or raw pointers (`this`) outlive background job execution.\n*   **Refcount Draining & Destruction Races**:\n    *   Verify that root jobs \u002F parent structures hold an explicit refcount credit during worker dispatch so refcounts cannot hit zero mid-execution on worker failure.\n\n### Pillar 5: Test Integrity & False-Pass Audits\n*   **Tautological Assertions**:\n    *   Inspect test assertions to ensure they are not tautologies (e.g., assertions that can never evaluate to false).\n*   **Concurrency Stress & Race Coverage**:\n    *   Verify that multi-threaded tests genuinely create contention on the guarded path rather than executing deterministically or passing due to timing coincidence.\n*   **Build Wiring**:\n    *   Verify new test files are wired into the corresponding `CMakeLists.txt`.\n\n### Pillar 6: Architectural Completeness & Scaling\n*   **Feature Completeness**:\n    *   When introducing a new capability (e.g., adopted worker threads), verify that all sizing, partitioning, and helper calculations (e.g., `parallel_for` helper job counts) scale with the new capability rather than only querying legacy state.\n\n---\n\n## 3. Standard Review Output Format\n\nReviews must be structured using the following format:\n\n```markdown\n\u003CFile Path>\n  - \u003Cline> [\u003Ccategory>] \u003CConcise finding description detailing the exact failure mechanism.>\n\n...\n\nReview Summary:\n- **The Blocker**: Identify the single most critical blocking issue, explaining the thread interleaving or failure sequence.\n- **Intent vs. Code**: Assess whether the PR achieves its architectural goals and note any systematic omissions.\n- **Explicitly Checked & Ruled Out**: List non-obvious complex areas investigated and verified as safe\u002Fcorrect (proves thorough inspection and eliminates false positives).\n- **Minor Cleanups**: Note dead code, minor un-decremented tracking counters, or triplicated logic below the blocking finding bar.\n```\n\n### Category Tags:\n*   `[correctness]` — Logic errors, data races, memory ordering bugs, lost wakeups, narrowing bugs, crashes.\n*   `[efficiency]` — Scaling bottlenecks, false sharing, unnecessary allocations, missed parallelism.\n*   `[api-contract]` — Broken traits, dropped interface support, violated preconditions\u002Fpostconditions.\n*   `[behavior-change]` — Silent changes to edge-case behavior (e.g., `count == 0`).\n*   `[test-coverage]` — Tautological tests, missing failure checks, unwired tests.\n\n---\n\n## 4. Review Execution Workflow for AI Agents\n\nWhen instructed to review code, a branch, a commit range, or a PR:\n\n1. **Discover Changes**:\n   ```bash\n   git diff \u003Cbase_branch>...\u003Ctarget_branch>\n   # or\n   git log -n \u003Ccount> --stat\n   ```\n2. **Read Related Context**:\n   Read the modified files alongside their callers and dependencies using `view_file` to understand the complete cross-function state machine.\n3. **Execute the 6 Analytical Pillars**:\n   Systematically audit the diff against each pillar.\n4. **Draft & Filter Findings**:\n   Verify every potential finding against the codebase to eliminate false positives.\n5. **Format Report**:\n   Output the structured findings following the format in Section 3.\n",{"name":5,"description":7},"main","fd9ec7ff9510df148011c5326bcb50d4bfb811ca","skills\u002Fcode_review",2305,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament","https:\u002F\u002Fgoogle.github.io\u002Ffilament\u002F","https:\u002F\u002Fgithub.com\u002Fgoogle","https:\u002F\u002Fopensource.google\u002F","Google ❤️ Open Source","2026-11-14 13:54:43",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":71},[61,64,67],{"tool":62,"command":63},"npx","npx skills add google\u002Ffilament --skill code-review",{"tool":65,"command":66},"bunx","bunx skills add google\u002Ffilament --skill code-review",{"tool":68,"command":69},"pnpm","pnpm dlx skills add google\u002Ffilament --skill code-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament\u002Fblob\u002Fmain\u002Fskills\u002Fcode_review\u002FSKILL.md",true,[],[74,94,109,125,147,163],{"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":44},"01M698XTTZP1MWPYW9EVV7DF2Q","filament-math","Enforce correct mathematical primitives and naming conventions in Filament code. Use this skill when performing coordinate transformations, vector math, or projection setups.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament\u002Fblob\u002Fmain\u002Ftools\u002Fai\u002Fagents\u002Fskills\u002Ffilament_math\u002FSKILL.md",[],[81,82,83,31,84,85,86,39,87,88,89,90,91,92,93],"filament math","math","when","vector","correct","primitives","coordinate","enforce","conventions","Mathematical","naming","naming conventions","performing",{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":44},"01M698XRXR5Q463GQ03RDBD9FR","filament-lifetime","Enforce correct Filament object creation and destruction lifecycles. Use this skill when initializing rendering pipelines or allocating\u002Fdeallocating GPU buffers and entities.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament\u002Fblob\u002Fmain\u002Ftools\u002Fai\u002Fagents\u002Fskills\u002Ffilament_lifetime\u002FSKILL.md",[],[101,39,102,103,88,104,85,105,106,107,83,108],"filament lifetime","rendering","creation","pipelines","gpu","destruction","object","entities",{"skillNo":110,"name":111,"nameZh":6,"description":112,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":113,"categories":114,"keywords":115,"publishedAt":44},"01M698XQ0E9RE6RDFYKS2PF1S1","cherry-pick","Cherry-pick a commit onto another branch and open the PR. Use when the user asks to cherry-pick or backport a commit or PR to a release branch.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament\u002Fblob\u002Fmain\u002Fthird_party\u002Fdawn\u002Fthird_party\u002Fdirectx-shader-compiler\u002Fsrc\u002F.github\u002Fskills\u002Fcherry-pick\u002FSKILL.md",[],[116,117,118,119,83,120,39,121,122,123,124],"pick","cherry pick","branch","open","user","backport","commit","pr","release",{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":131,"publishedAt":44},"01M698XN2P44Y956SJRD7CPB22","verification-protocols","Standard verification pipeline to execute after modifying C++ source or header files. Use this skill to format includes, build the engine, and run core tests.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament\u002Fblob\u002Fmain\u002Fskills\u002Fverification_protocols\u002FSKILL.md",[],[132,133,134,135,136,137,138,139,140,39,141,142,143,144,145,146],"protocols","verification","verification protocols","core","run","pipeline","format","engine","execute","files","build","c++","tests","source","standard",{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":153,"publishedAt":44},"01M698XK6J428QZM1WGVWW50JE","preprocessor-guard-hygiene","Enforce preprocessor guard rules and macro-dependent inclusion hygiene in C++. Use this skill when modifying conditional imports or preprocessor directives.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament\u002Fblob\u002Fmain\u002Fskills\u002Fpreprocessor_guard_hygiene\u002FSKILL.md",[],[154,155,156,157,83,158,159,160,161,143,88,39,162],"hygiene","guard","preprocessor guard hygiene","Rules","directives","conditional","imports","inclusion","macro",{"skillNo":164,"name":165,"nameZh":6,"description":166,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":167,"categories":168,"keywords":169,"publishedAt":44},"01M698XHA1J01F915DKYHXZ18C","header-self-containment","Enforce strict header self-containment so that every header file compiles independently. Use this skill when creating or modifying C++ header files.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Ffilament\u002Fblob\u002Fmain\u002Fskills\u002Fheader_self_containment\u002FSKILL.md",[],[170,171,172,83,143,88,141,39,173],"header self containment","containment","self","file"]