Results for “theta”

Train, evaluate, export, quantize, and run inference for a Deformable DETR 2D object detection model using TAO, with deformable attention for efficient multi-scale feature processing.

Stress-test ideas, plans, and decisions using structured critical reasoning across five modes: Socratic questioning, dialectic, pre-mortem, red teaming, and evidence audit.

Implements finite automata, regular expressions, parsers, and lexers for building text processors and pattern matchers.

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques for 1.5-3.6× speedup without quality loss.

Measures the trade-off between computation time and energy consumption in mobile edge computing by computing a weighted sum of the two objectives, given system configuration parameters and per-user task characteristics.

Detect SSO and OAuth token replay and SaaS lateral movement using identity telemetry from Microsoft Entra ID and Okta.

Use this skill for dark mode tokens, contrast, theme persistence, system preference, color scheme QA. Trigger when the task involves web design work related to Dark Mode Theme System, implementation, audits, debugging, strategy, or validation.

`task-agent`/`review-agent`: use when broker delivery, ordering, acknowledgement, DLQ, backpressure, or replay changes; skip synchronous retry without message semantics.

Applies Schema.org structured data principles to guide decision-making, structure analysis, and improve search visibility outcomes.

Expert guidance for Xata, the serverless data platform that combines PostgreSQL, Elasticsearch, and AI capabilities in a single API. Helps developers build applications with full-text search, vector similarity search, file attachments, and branching — all through a type-safe TypeScript SDK.

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

Strata by Alpen Labs: ZK rollup on Bitcoin, BitVM-based bridge, EVM-compatible execution. Emerging Bitcoin L2 in 2025-2026. USE WHEN: evaluating Strata for deployment, comparing with Citrea.

Inspects metadata, hidden channels, and appended payloads in media files to recover concealed data in steganography challenges.

Create data-driven charts with Vega-Lite and Vega, covering bar, line, scatter, heatmap, area, radar, and word cloud visualizations from structured data arrays.

Configure, operate, and master KEDA (Kubernetes Event-driven Autoscaling) — ScaledObject, ScaledJob, TriggerAuthentication CRDs, 70+ scalers, HPA behavior tuning, scale-to-zero, the KEDA HTTP Add-on, production hardening, multi-trigger semantics, scalingModifiers formulas, GitOps integration, and troubleshooting stuck scalers. Covers the common traps (cooldownPeriod only applies to N→0, CPU/memory cannot drive scale-to-zero alone, activationThreshold vs threshold, multi-trigger max-of semantics, HPA conflicts).

Provides a rigorous thinking partner profile that challenges assumptions, calibrates claims to evidence, and improves decision quality under uncertainty.

Mines a codebase for evidence-backed improvement opportunities, forcing two escalation gates to surface breakthrough ideas, and outputs a ranked portfolio with a plan seed without implementing.

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

Provides a comprehensive design system specification inspired by The Verge's 2024 redesign, including color palette, typography, and component styles for building dark-themed editorial interfaces.

Selects the right Gemma model for a task, recommends deployment tooling (Gradio, Transformers.js, Vertex AI, MLX), and applies optimizations like MTP and QAT.

Intercepts responses to let users choose the depth and token budget before answering, with heuristic token estimation and preset depth levels.

Stack the Tech: the right tech and the right signals

Price-to-win lens using GSA CALC+, BLS OEWS, and incumbent USASpending award patterns for a pursuit. Use when user asks for realism checks or competitive pricing posture before proposal — draft skill, not production-verified.

Predict biomolecular complexes with Chai-1, an open AlphaFold3-style model that folds multi-entity assemblies (proteins, ligands, nucleic acids) from a single typed FASTA — strong on antibody–antigen and protein–ligand complexes, with optional MSA and restraint inputs. Use when predicting an antibody–antigen complex, folding a mixed protein/ligand/nucleic-acid assembly described in one FASTA, or generating a complex with experimental restraints. For binding-affinity prediction or a ligand-focused co-fold prefer alterlab-boltz; for protein-only or protein–protein folding prefer alterlab-alphafold; to dock into a fixed receptor prefer alterlab-diffdock. Part of the AlterLab Academic Skills suite.

Llama 3: The Llama 3 Herd of Models

Defines the LEND.AI assistant persona: identity, tone, and interaction rules for a Spanish-speaking senior mentor that teaches while working, questions decisions, and requires confirmation before acting.

Guiding workflows by decomposing complex tasks (Epics) into Atomic Steps under 15 minutes each, with progress tracking and drift prevention. Use when complex decomposition is needed.

【河图】SKILL.md — 全术数技能系统 v2.0

Documents Meta's design system for commerce surfaces, including color tokens, typography hierarchy, spacing, and responsive patterns for hardware merchandising.

Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups. Use when reproducing OpenVLA-OFT paper results, training custom VLA action heads (L1 or diffusion), deploying server-client inference for ALOHA, or debugging normalization, LoRA merge, and cross-GPU issues.

Universal deep research agent team. 13-agent pipeline for rigorous academic research on any topic. 7 modes: full research, quick brief, paper review, lit-review, fact-check, Socratic guided research dialogue, and systematic review with optional meta-analysis. Covers research question formulation, Socratic mentoring, methodology design, systematic literature search, source verification, cross-source synthesis, risk of bias assessment, meta-analysis, APA 7.0 report compilation, editorial review, devil's advocate challenges, ethics review, and post-research literature monitoring. Triggers on: research, deep research, literature review, systematic review, meta-analysis, PRISMA, evidence synthesis, fact-check, guide my research, help me think through, 研究, 深度研究, 文獻回顧, 文獻探討, 系統性回顧, 後設分析, 事實查核, 引導我的研究, 幫我釐清, 幫我想想, 我不確定要研究什麼, 研究方向, 研究主題.

Structured threat modeling skill using the PASTA framework (Process for Attack Simulation and Threat Analysis) combined with Adam Shostack's 4-question framework. Use this skill whenever the user asks to do threat modeling, security analysis, map the attack surface, identify threats, or review an application for security risks — even if they don't mention PASTA or a specific framework by name. Core activities: Component Mapping (architecture + data flows), Critical Assessment (business impact prioritization), and Logic Flaw Identification (attacker mindset on business logic). Produces: component map diagram (Mermaid), data flow diagram (Mermaid), attack tree (Mermaid), STRIDE threat table, prioritized risk register, and an actionable mitigation plan. Invoke proactively for any security review, architecture review, or "what could go wrong?" session.

Runs and writes tests, verifies behavior, and orchestrates two-pass TDD across scoped modes, blocking on quality gates and never editing production code.

Generates at most one token recommendation and one prediction-market pick per run, each with a falsifiable thesis, entry, sizing, and kill criterion. Returns NO_PICK when no candidate meets the bar.

Surface the most valuable next action by combining task stack, queue state, inbox pressure, health, and goals. Recommends one specific action with rationale. Triggers on "/next", "what should I do", "what's next".