Results for “guardrails”

Use to design support incentives that improve behaviour without destroying the metric — pairing pay with guardrails, naming gaming modes, and choosing measures that survive Goodhart pressure. Trigger for "incentive plan", "agent bonus scheme", "SPIFF design", "pay for QA score", "what metric should we bonus", CSAT incentives, or reviewing whether a comp change is driving gaming.

Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, LEACE, SAE decomposition, etc.) to excise guardrails while preserving reasoning. 9 CLI methods, 28 analysis modules, 116 model presets across 5 compute tiers, tournament evaluation, and telemetry-driven recommendations. Use when a user wants to uncensor, abliterate, or remove refusal from an LLM.

This skill should be used when the user asks to "copy trades from" a wallet, "mirror a wallet", "follow this address", set up "copy trading", "track and replicate a trader", or mirror another account's swaps bounded by guardrails. Watches a target wallet and mirrors its trades, filtered by chain, asset match, position size, and the follower's own portfolio state.

TIGER v2 — Multi-scanner trading system for Hyperliquid perps via Senpi MCP. 5 signal patterns (BB compression breakout, BTC correlation lag, momentum breakout, mean reversion, funding rate arb), DSL v4 trailing stops, goal-based aggression engine, and risk guardrails. Configurable profit target over deadline. 12-cron architecture (10 TIGER + prescreener + ROAR meta-optimizer). Pure Python analysis. Requires Senpi MCP, python3, mcporter CLI, and OpenClaw cron system.

Caching strategies across the RAG stack. Semantic caching with GPTCache and LangChain, Redis-based embedding-similarity cache, cache key design, TTL/invalidation, partial caching (cache retrieval only), provider-native prompt caching (Anthropic, OpenAI), and hierarchical L1/L2 caches. USE WHEN: user mentions "semantic cache", "GPTCache", "LLM cache", "prompt caching", "Redis vector cache", "cache invalidation for RAG", "reduce LLM cost", "latency reduction LLM" DO NOT USE FOR: retrieval accuracy - use `rag-patterns`; groundedness checks - use `rag-guardrails`; incremental indexing - use `rag-production`

Knowledge-graph-augmented retrieval. Entity and triple extraction, graph construction (Neo4j, LlamaIndex PropertyGraphIndex), hierarchical community summarization (Microsoft GraphRAG), personalized PageRank (HippoRAG), multi-hop traversal retrieval, and hybrid graph + vector pipelines. USE WHEN: user mentions "GraphRAG", "HippoRAG", "knowledge graph RAG", "entity extraction", "multi-hop reasoning", "Neo4j RAG", "LlamaIndex property graph", "LangChain graph retriever", "triple extraction", "community summarization" DO NOT USE FOR: vanilla vector RAG - use `rag-patterns`; multimodal inputs - use `multimodal-rag`; production indexing ops - use `rag-production`; hallucination checks - use `rag-guardrails`

Route LangSmith work into one workflow packet before touching SDK code. Use when the user needs LangSmith tracing, offline evals, annotation/review queues, prompt-registry decisions, audit/gap review, or cross-service trace propagation for an LLM app or agent workflow. Choose one packet: trace-debug, eval, review, prompt-registry, propagation, or audit. Triggers on: LangSmith, LangChain tracing, `@traceable` / `traceable`, `wrap_openai` / `wrapOpenAI`, datasets, experiments, annotation queues, feedback criteria, Prompt Hub, run trees, trace IDs, or production confidence for an AI feature. Not for generic SLO/alert design, non-LangSmith deployment orchestration, or runtime guardrails outside LangSmith.

Write a rigorous, decision-grade experiment spec — falsifiable hypothesis, primary metric, guardrails, randomisation unit, exposure definition, method (A/B, holdout, switchback, quasi-experiment, MAB), MDE/duration plan, peek policy, validity threats, and pre-committed decision rule. Platform-agnostic. Load when the user has a candidate experiment and needs to spec it before launch, or says "spec this experiment", "write the test plan", "design this A/B test", "what's the hypothesis", "how big a sample do we need", "how long should we run this", "define the metrics for this test", or when the experimentation orchestrator routes here.

Security controls for RAG. Indirect prompt-injection via retrieved documents, PII detection/redaction (Microsoft Presidio, AWS Comprehend), multi-tenant isolation, ACL-aware retrieval with row-level/metadata filtering, data-leakage prevention, jailbreak hardening on retrieved context, GDPR right-to-be-forgotten in vector DBs. USE WHEN: user mentions "prompt injection RAG", "indirect prompt injection", "PII redaction", "Presidio", "ACL RAG", "row-level security", "multi-tenant RAG isolation", "GDPR vector DB", "right to be forgotten", "jailbreak", "data leakage RAG" DO NOT USE FOR: hallucination detection - use `rag-guardrails`; tenancy scaling patterns - use `rag-production`; audit tracing schema - use `rag-observability`

Drafts litigation-ready U.S. pre-suit demand letters that function as settlement instruments and defensible future exhibits. Enforces element-driven narratives, verified authority, damages methodology, and ethics guardrails. Use this skill when the user mentions demand letter, pre-suit demand, breach and cure notice, settlement demand, insurance policy-limits demand, Stowers demand, FDCPA collection letter, notice of intent to file, cease and desist demand for payment, or pre-litigation correspondence. Also trigger when the user asks about FRE 408 framing, contractual notice compliance, statutory pre-suit prerequisites, evidence preservation notices, or quantifying damages for a demand package. Skill includes 14 verified real-letter exemplars referenced from `references/ARCHETYPE-INDEX.md`.

Set up pre-commit hooks on a Python project — standard file-hygiene checks plus a security gate (gitleaks secret scanning, Trivy filesystem scan for CVEs/secrets/misconfigs, and Bandit Python SAST). Use this whenever the user wants to add, configure, or fix pre-commit hooks on a Python repo, mentions .pre-commit-config.yaml, wants secret/vulnerability/SAST scanning on commits, or is setting up code-quality guardrails — even if they just say 'add pre-commit hooks' without naming the tools. Especially for uv-based GenAI/LLM backends. Handles the setup gotchas that break first-time installs: the Trivy binary, the required data/html.tpl report template, bandit[toml] + [tool.bandit] config, and the right .gitignore entries.

Run a structured retrospective after development-phase runs of your product's agents — interview the owner in plain language about what went well and poorly, draft ranked improvement hypotheses, then design and run small n=1/n=2 experiments with pre-declared success criteria, guardrails, stop conditions, and a cost/ROI kill-switch. Load when the user says how did that run go, retro this run, the agent output was bad, what should we improve, draft hypotheses, run a small experiment, or after repeated dev runs of an agentic system produce uneven quality. Priority: output quality over performance over cost, each with diminishing-returns stops. NOT a product A/B test (experimentation), NOT coding-agent harness repair (harness-evolution), NOT production-scale learning (runtime-learning-loop).