Agent Skills

by posthog

Follow-up scout for the Signals inbox itself. Watches reports that recently transitioned to resolved (an implementation PR merged) and, after a deployment soak window, re-measures the underlying problem to check the fix actually held — plus a strictly-gated escalation check on recently dismissed reports. Emits findings only when a shipped fix demonstrably didn't hold; confirmations and unverifiable verdicts become durable memory and an empty close-out. Self-contained peer in the signals-scout-*

signals-scout-logs

by posthog

Focused Signals scout for PostHog projects using logs. Watches for volume bursts, severity-distribution shifts, service silence, fresh message patterns, and trace-correlated bursts via the logs ingestion pipeline. Emits findings only when they clear the confidence bar; otherwise writes durable memory and closes out empty. Self-contained peer in the signals-scout-* fleet — no dependencies on other skills.

signals-scout-revenue-analytics

by posthog

Focused Signals scout for PostHog projects using revenue analytics. Watches the derived revenue product for upstream failures (Stripe sync stalls, capture regressions), config drift (missing subscription property, currency mix surprises, broken Stripe↔person joins, deferred-revenue gaps), and goal-miss escalations. Emits findings only when they clear the confidence bar; otherwise writes durable memory and closes out empty. Self-contained peer in the signals-scout-* fleet — no dependencies on oth

signals-scout-ai-observability

by posthog

Focused Signals scout for PostHog projects using AI observability. Rotates through a set of lenses — cost, latency, errors, volume, eval performance, eval/enrichment config, clusters, and tool usage — watching each for trends and spikes sliced by the dimensions it discovers over time. Leans on the sandbox's bundled `exploring-llm-*` deep-dive skills for the actual queries. Emits findings only when they clear the confidence bar; otherwise writes durable memory and closes out empty. Self-contained

signals-scout-error-tracking

by posthog

Focused Signals scout for PostHog projects using error tracking. Watches `$exception` bursts, stuck loops, multi-fingerprint clusters, status regressions, and stack-trace activity-name patterns. Emits findings only when they clear the confidence bar; otherwise writes durable memory and closes out empty. Self-contained peer in the signals-scout-* fleet — no dependencies on other skills.

signals-scout-surveys

by posthog

Focused Signals scout for PostHog projects running surveys. Watches active surveys for score regressions (NPS / CSAT / rating drops), response-volume drops, abandonment spikes, and targeting drift, AND aggregates open-text responses into recurring themes the team should know about (clusters of complaints, praise, feature requests). Emits findings only when a theme or anomaly clears the confidence bar; otherwise writes durable memory and closes out empty. Self-contained peer in the signals-scout-

signals-scout-experiments

by posthog

Focused Signals scout for PostHog projects running A/B experiments. Watches running experiments for validity threats (sample ratio mismatch, multi-variant contamination, exposure stalls, mid-run flag mutations) and lifecycle drift (zombie experiments running long past their useful life, decided-but-still-running experiments, ended experiments whose flags still serve multiple variants). Emits findings only when they clear the confidence bar; otherwise writes durable memory and closes out empty. S

signals-scout-csp-violations

by posthog

Focused Signals scout for PostHog projects collecting Content Security Policy (CSP) violation reports. Watches `$csp_violation` events for fresh blocked-URL clusters, per-directive bursts, page-scoped regressions after deploys, and suspicious third-party domains that may indicate a compromised script. Emits aggregated findings only when a cluster clears the confidence bar; otherwise writes durable memory and closes out empty. Self-contained peer in the signals-scout-* fleet — no dependencies on

llama-cpp

🔥🔥🔥✓ VerifiedFreeAdvanced setup

by firecrawl

Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.

sentencepiece

🔥🔥🔥✓ VerifiedFreeNeeds API keys

by firecrawl

Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or reproducible tokenization.

awq-quantization

by firecrawl

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

crewai-multi-agent

by firecrawl

Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for production workflows requiring sequential/hierarchical execution. Built without LangChain dependencies for lean, fast execution.

optimizing-attention-flash

by firecrawl

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when training/running transformers with long sequences (>512 tokens), encountering GPU memory issues with attention, or need faster inference. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding window attention.

langchain

🔥🔥🔥✓ VerifiedFreeNeeds API keys

by firecrawl

Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications. Best for rapid prototyping and production deployments.

gptq

🔥🔥🔥✓ VerifiedFreeNeeds API keys

by firecrawl

Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.

pytorch-fsdp2

by firecrawl

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.

unsloth

by firecrawl

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

serving-llms-vllm

by firecrawl

Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.

peft-fine-tuning

by firecrawl

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

quantizing-models-bitsandbytes

by firecrawl

Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.