性能
性能分析与优化
cross-conversation-project-manager
gked2121
Maintain project state across MULTIPLE conversations over days/weeks. Track tasks, decisions, blockers, resources. Auto-update when project mentioned. Generate status reports and proactive reminders.
serving-llms-vllm
ariffazil
"vLLM: high-throughput LLM serving, OpenAI API, quantization."
lookalike-customer-finder
gked2121
Input your best customers and find 100+ companies that match the profile. Uses firmographic data, tech stack, growth signals, and similarity scoring to identify ideal prospects. Use when building target account lists or expanding to new markets.
hermes-system-audit
joecastelino
Run a comprehensive system audit of the AMG Hermes fleet — gateway services, persistent browser servers, cron jobs, GBrain, disk, memory, and agent profiles. Use when Joe reports agent degradation, after weekend incidents, or for recurring health checks.
context-manager
Toowiredd
Manages permanent memory storage for decisions, blockers, context, preferences, and procedures. Use when user says "remember", "save this decision", "what did we decide", "recall", "search memories", "any blockers", or when making important architectural decisions. Provides SDAM compensation through external memory.
session-launcher
Toowiredd
Restores full context when user says "hi-ai" or starts a new conversation. Searches project files, loads memory indexes, reads session state, and creates visual dashboard showing current project, recent decisions, active blockers, and quick actions. Use when user says "hi-ai", "continue", "restore context", or starts a fresh conversation.
python-performance-optimization
spjoshis
Profile and optimize Python code using cProfile, memory profilers, and performance best practices. Use when debugging slow Python code, optimizing bottlenecks, or improving application performance.
unit-economics-optimizer
famaoai-creator
Output file path
performance-audit
truongnat
Audit and optimize application performance across frontend, backend, and database layers. Identifies bottlenecks, provides actionable optimization recommendations, and measures improvements. Use when the app feels slow, Lighthouse scores are low, or API response times exceed targets.
rl-newton
Idate96
"End-to-end Newton RL workflow in moleworks_newton: local smoke, cluster submit/monitor/debug, sync logs, benchmark checkpoints, and run post-benchmark analysis with strict experiment ledgers."
metering
SpiralCloudOmega
Audio metering — peak, RMS, VU, EBU R128 loudness, spectral analyzer, correlation meter
backend-principle-eng-cpp-pro-max
PrakharMNNIT
"Principal backend engineering intelligence for C++ systems and performance-critical services. Actions: plan, design, build, implement, review, fix, optimize, refactor, debug, secure, scale backend code and architectures. Focus: correctness, memory safety, latency, reliability, observability, scalability, operability."
backend-principle-eng-javascript-pro-max
PrakharMNNIT
"Principal backend engineering intelligence for JavaScript services. Actions: plan, design, build, implement, review, fix, optimize, refactor, debug, secure, scale backend code and architectures. Focus: correctness, reliability, performance, security, observability, scalability, operability, cost."
ask-questions-if-underspecified
alemar11
Clarify requirements before implementing when a request is underspecified or the user asks for clarification.
system-scenario-player
famaoai-creator
Global speed multiplier for delays.
sakthai-memory-recall
beer-sakthai
Recall relevant memory before answering.
context-builder
molotroka123-cell
Assemble the smallest high-quality task context from memory, semantic code search, current files, tool results and user instructions while protecting the model context window from irrelevant material.
sakthai-memory-consolidate
beer-sakthai
Periodically fold and dedupe memory to keep it sharp.
sprint-planning
spjoshis
Master sprint planning with capacity calculation, story selection, commitment, and effective planning techniques for successful sprints.
SakJules-cycle-growth
beer-sakthai
"Fold the cycles lessons back into memory and skills."
llm-inference
eyadsibai
Use when "LLM inference", "serving LLM", "vLLM", "llama.cpp", "GGUF", "text generation", "model serving", "inference optimization", "KV cache", "continuous batching", "speculative decoding", "local LLM", "CPU inference"
ask-questions-if-underspecified
ederheisler
Clarify requirements before implementing. Use when serious doubts araise.
perf
coreindustries
"Profile, benchmark, and optimize application performance."
memory-systems
eyadsibai
Use when implementing agent memory, persisting state across sessions, building knowledge graphs, tracking entities, or asking about "agent memory", "knowledge graph", "entity memory", "vector stores", "temporal knowledge", "cross-session persistence"