observability-architecture
LiteLLM-RS Observability Architecture. Covers Prometheus metrics, OpenTelemetry tracing, structured logging, health checks, and alerting integration.
Découvrez les skills Outils, comparez leurs dépôts et consultez leurs instructions.
LiteLLM-RS Observability Architecture. Covers Prometheus metrics, OpenTelemetry tracing, structured logging, health checks, and alerting integration.
LiteLLM-RS Provider 开发与架构指南。用于添加新 provider、迁移错误处理、维护 66+ provider 的一致性。包含架构对比分析和最佳实践。
LiteLLM-RS Error Handling Architecture. Covers two-tier error hierarchy, ProviderError factory methods, HTTP status mapping, retry logic, and error context preservation.
LiteLLM-RS Authentication Architecture. Covers JWT + API Key + RBAC multi-method auth, rate limiting with DashMap, middleware pipeline, and secure credential management.
LiteLLM-RS Streaming Architecture. Covers UnifiedSSEParser, SSETransformer trait, VecDeque buffering, provider-specific transformers, and real-time event handling.
LiteLLM-RS Caching Architecture. Covers Redis caching, vector database semantic caching, multi-tier cache strategy, TTL management, and cache invalidation patterns.
LiteLLM-RS Configuration Architecture. Covers YAML loading, environment variable override, validation patterns, type-safe config models, and hot reloading.
Split large tasks into smaller subtasks. Use when a task spans many modules or requires phased implementation.
Execute next task from Implementation Plan. Use when user asks to implement next task, start next feature, or continue implementation.
Add or modify tasks in Implementation Plan. Use when user proposes new requirements, changes priorities, or adjusts existing task scope.
Maintain sloc-guard project documentation. Use when user completes a feature and asks to update docs.
Execute next task from Implementation Plan. Use when user asks to implement next task, start next feature, or continue implementation.
Get the current Neovim context as JSON (cursor position, current file, visual selection and diagnostics) to help answer questions about code at the current cursor position, visual selections and diagnostics. Us...
Utility layer for Kysely on SQLite databases. Use when working with SQLite via Kysely for: schema definition and auto-migration, soft-delete patterns, precompiled queries, pagination, nested transactions using...
Analyze checkpoint tests and suggest missing edge cases. Use after writing tests or when reviewing test coverage. Invoke with /edge-cases .
Analyze checkpoint tests and suggest missing edge cases. Use after writing tests or when reviewing test coverage. Invoke with /edge-cases .
Analyze test cases against specifications to find ambiguous assumptions. Use when tests might be making assumptions not explicitly defined in the spec. Invoke with /test-ambiguity-detector .
Validate all checkpoints from an agent run directory in parallel. Spawns test-validator agents for each checkpoint and summarizes results. Invoke with /validate-run [problem].
Run problem tests using eval-snapshot instead of raw pytest. Use this to evaluate solutions against benchmark tests in Docker. Invoke with /run-tests .
Reclassify tests by adding @pytest.mark.functionality to tests not explicitly shown in the spec. Invoke with /reclassify-tests .
Audit a checkpoint specification for realism and design decision forcing. Reviews specs to remove hand-holding, hidden corner cases, and architectural giveaways. Invoke with /audit-spec .
Continually test and repair solutions to benchmark problems until they pass. Use when a solution has failing tests. Invoke with /fix-solution .
Review checkpoint specs and tests to identify tests that encode ambiguous interpretations rather than explicit requirements. Use when asked to check checkpoint_N.md against test_checkpoint_N.py, when auditing t...
Guide for adding new files to this codebase while respecting architectural principles including Separation of Concerns, Common Closure Principle, small composable functions (max 20 lines), and externalizing con...
Le classement utilise les étoiles du dépôt GitHub indexé. Plusieurs skills peuvent partager ce nombre : ce n’est ni un nombre d’installations ni une note de qualité.