file-operations retrieves file metadata, line counts, and content statistics via shell commands and grep patterns. covers single and batch analysis without modifying source files.

sentry-node-sdk guides setup of error monitoring, tracing, and profiling across node.js, bun, and deno runtimes. includes framework-specific integration paths for express, fastify, koa, and others with version guardrails.

sentry-cloudflare-sdk covers invocation rules for Sentry setup across Cloudflare Workers, Pages, and associated runtimes, but lacks procedural depth and failure handling.

ziniao-department covers crud operations on organizational units including hierarchical tree display, creation with parent constraints, cascading deletion, and reordering. references external docs for command details and shared auth/error handling rules.

meticulous-simulate-and-diff runs a session replay against a live url and optionally compares screenshots to a baseline replay. covers simulation setup, prerequisite checks, and diff interpretation without deep guidance on failure modes or result analysis.

meticulous-review walks through visual inspection of test diffs via CLI and MCP tools, comparing screenshots and DOM changes to classify frontend modifications as intended or unexplained.

primevue covers installation, plugin configuration, auto-import setup, and theme selection for vue 3 and vite projects. includes common component examples and preset theme options.

deep-research orchestrates multi-step autonomous research via gemini api, planning and synthesizing information into reports. covers market analysis, competitive landscaping, and technical literature reviews with configurable output formats.

eas-hosting covers deploying expo web apps and api routes to cloudflare workers via eas deploy. covers paid hosting model, api route authoring decisions, and basic deployment workflow.

guidance for diagnosing and fixing unhandled errors from vs code telemetry by tracing malformed data to its source rather than patching crash sites.

smux provides tmux pane control and cross-agent message passing via a tmux-bridge CLI. it abstracts away polling and blocking, routing replies directly back to the calling agent through tmux pane channels.

detects new or modified skills in .agents/skills/ and publishes them to ai-skills after git validation, with snapshot capability for rollback and manual review before deployment.

vtex-io-client-integration covers selecting and configuring client libraries for backend service calls in vtex io apps, including native client choice, custom client creation, and instance option tuning.

creating-secrets-using-best-practices wraps AWS Secrets Manager operations with KMS encryption, rotation policies, and least-privilege IAM controls. covers four secret types and includes lifecycle management guardrails.

angular-migration covers hybrid app setup, component and service conversion patterns, and dependency injection interop between angularjs and angular. lacks step-by-step migration sequencing and failure recovery guidance.

react-best-practices aggregates 45 performance rules across 8 categories for react and next.js, organized by priority and impact. references external rule files for detailed explanations but lacks inline procedures.

agent-email-inbox provides webhook-based email ingestion patterns for ai agents, emphasizing security and event-driven design over polling. covers inbox setup, content safety, and webhook architecture but lacks complete procedural depth.

design-system covers three-layer token architecture (primitive/semantic/component), css variable generation, component state specs, and slide generation with contextual decision logic. includes scripts for token validation and bm25-based slide search.

go-linting establishes consistent linting across go projects via golangci-lint configuration, with a philosophy prioritizing uniformity over prescriptive rules.

mysql-patterns outlines activation triggers and version-checking prerequisites for mysql/mariadb schema, migration, query tuning, and replication work, but truncates before exposing concrete procedures.

openhue controls philips hue lights, rooms, and scenes via cli commands. supports brightness, color, temperature, and preset modes for local bridge-based automation.

agentmemory-architecture describes the iii-engine design of a local memory server for agents, covering hybrid retrieval, storage lifecycle, and network ports, but lacks procedural depth and actionable guidance.

edge-concept-synthesizer clusters detector tickets into abstracted concepts with thesis and invalidation signals, enabling concept-level review before strategy implementation without direct overfitting.

waapi enables synchronous keyframe animation control via web animations api seeking, integrated with hyperframes timeline positioning and deterministic state persistence.

components-guide covers convex components as isolated backend modules bundling schema and functions. explains sibling pattern, when to use components over monolithic code, and constraints around data isolation and cross-component access.

dingtalk-document provides structured guidance for workspace and document operations via dingtalk api. workflow emphasizes configuration validation and helper script delegation, with references deferred to external api.md.

react-best-practices guides component design through a decision tree favoring render-time logic and event handlers over effects, with emphasis on effects as external system synchronization only.

product-marketing guides users through creating and updating a stored product marketing context document, checking for existing files and helping populate positioning and messaging sections incrementally.

git-guardrails-claude-code installs a PreToolUse hook to intercept and block dangerous git commands (push, reset --hard, clean, branch -D, checkout) at the project or global scope, preventing destructive operations in claude code execution.

session-start-hook provides structured input schema and event wiring for Claude Code initialization, enabling dependency installation across CLI/web/API contexts. Execution logic and recovery paths remain incomplete.

agentic-actions-auditor provides conceptual framing for auditing github actions workflows that invoke ai coding agents, but omits concrete procedural steps and concrete examples of detected vulnerabilities.

tailwind-css-patterns provides common utility compositions and responsive layouts for tailwind v4.1+, with code examples and troubleshooting guidance but lacks clear decision trees and outcome validation.

nelson orchestrates multi-agent task execution via a naval metaphor, enforcing structured mission planning with role boundaries, progress tracking, and decision logging across sequential or parallel workflows.

figma-use establishes the mandatory prerequisite workflow for executing javascript against figma files via the plugin api. it prescribes skill loading order and tool batching patterns but lacks detailed procedural steps and failure mode coverage.

hot-topics fetches real-time trending data from six major chinese platforms (weibo, zhihu, baidu, douyin, toutiao, bilibili) via a single unified api. includes platform-specific examples, caching advice, and use cases for trend analysis and comparison.

test-coverage identifies and remediates coverage gaps in flows apps to meet the mandatory 80% line coverage threshold, starting with framework verification and coverage report analysis.

writing-skills provides decision tree guidance and tiered templates for creating and improving agent skills, with emphasis on architecture patterns and pre-deployment compliance checks.

notion-automation provides templated workflows for database sync, form ingestion, cross-platform integrations, and content pipelines. covers 5 core workflows with api patterns and crm/project/content database structures, but lacks operational robustness around error handling and rate limits.

fusion-issue-authoring routes issue creation through type-specific agent modes (bug, feature, user-story, task, devils-advocate) and enforces shared safety gates before github mutation.

multi-agent-patterns covers supervisor, peer-to-peer, and hierarchical coordination models for distributing llm work across isolated context windows, emphasizing context partitioning over role anthropomorphization.

ux-copy provides structured templates and principles for writing interface microcopy across common patterns (buttons, errors, empty states, tooltips). it emphasizes clarity and user context but lacks procedural depth and failure mode guidance.

iterate-pr monitors ci status and review feedback on pull requests, with bundled scripts for fetching checks, categorizing comments, and posting replies. lacks explicit decision logic and recovery paths.

adk-eval-guide documents the evaluation framework and workflow for adk skills, referencing auxiliary guides for metrics, user simulation, and multimodal testing without spelling out the actual methodology.

ce-optimize runs metric-driven iterative loops to converge on measurable improvements. covers goal definition, measurement scaffolding, and parallel experiment orchestration across multiple platform harnesses.

rust-skill-creator automates skill generation from rust documentation by detecting execution mode (agent or inline) and executing appropriate workflows for crates, standard library, or custom urls.

remove-ai-style detects and rewrites ai-generated patterns across chinese and english text at three intensity levels (moderate, heavy, full), relying on reference documents and manual review to reduce formulaic language and restore natural voice.

hyperframes-read-first routes video creation requests to appropriate sub-skills via a capability map and documents the composition authoring contract, but lacks step-by-step procedures and explicit failure handling.