benchflow-ai/skillsbench - Page 2

audiobook

Create audiobooks from web content or text files. Handles content fetching, text processing, and TTS conversion with automatic fallback between ElevenLabs, OpenAI TTS, and gTTS.

elevenlabs-tts

ElevenLabs Text-to-Speech API for high-quality speech synthesis.

gtts

Google Text-to-Speech (gTTS) for converting text to audio. Use when creating audiobooks, podcasts, or speech synthesis from text. Handles long text by chunking at sentence boundaries and concatenating audio segments with pydub.

openai-tts

OpenAI Text-to-Speech API for high-quality speech synthesis. Use for generating natural-sounding audio from text with customizable voices and tones.

attention-variants-from-papers

Implement paper-defined attention variants by extracting mechanism invariants, mapping tensor shapes, preserving module interfaces, validating numerical behavior, and integrating the custom attention into an existing transformer stack.

modal-gpu

Run Python code on cloud GPUs using Modal serverless platform. Use when you need A100/T4/A10G GPU access for training ML models. Covers Modal app setup, GPU selection, data downloading inside functions, and result handling.

fragmentation-aware-packing

Choose placements that preserve useful residual capacity. Use for bin packing, GPU sharing, accelerator placement, and multi-resource scheduling where stranded capacity hurts future fit.

multi-resource-allocation-validation

Validate and repair proposed resource allocations by replaying them against temporary capacity. Use when actions consume several resource dimensions such as CPU, memory, GPUs, or accelerators.

online-resource-scheduling

Design deterministic online scheduling policies from current observations. Use when assigning arriving work to limited resources without seeing future requests.

mhc-algorithm

Implement mHC (Manifold-Constrained Hyper-Connections) for stabilizing deep network training. Use when implementing residual connection improvements with doubly stochastic matrices via Sinkhorn-Knopp algorithm. Based on DeepSeek's 2025 paper…

modal-gpu

Run Python code on cloud GPUs using Modal serverless platform. Use when you need A100/T4/A10G GPU access for training ML models. Covers Modal app setup, GPU selection, data downloading inside functions, and result handling.

nanogpt-training

Train GPT-2 scale models (~124M parameters) efficiently on a single GPU. Covers GPT-124M architecture, tokenized dataset loading (e.g., HuggingFace Hub shards), modern optimizers (Muon, AdamW), mixed precision training, and training loop implementation.

constraint-parser

Parse scheduling constraints from a email text.

gmail-skill

Manage Gmail - send, read, search emails, manage labels and drafts. Use when user wants to interact with their Gmail account for email operations.

automatic-speech-recognition-asr

Transcribe audio segments to text using Whisper models. Use larger models (small, base, medium, large-v3) for better accuracy, or faster-whisper for optimized performance. Always align transcription timestamps with diarization segments for accurate…

multimodal-fusion-for-speaker-diarization

Combine visual features (face detection, lip movement analysis) with audio features to improve speaker diarization accuracy in video files. Use OpenCV for face detection and lip movement tracking, then fuse visual cues with audio-based speaker embeddings.…

speaker-clustering-methods

Choose and implement clustering algorithms for grouping speaker embeddings after VAD and embedding extraction. Compare Hierarchical clustering (auto-tunes speaker count), KMeans (fast, requires known count), and Agglomerative clustering (fixed clusters). Use…

voice-activity-detection-vad

Detect speech segments in audio using VAD tools like Silero VAD, SpeechBrain VAD, or WebRTC VAD. Use when preprocessing audio for speaker diarization, filtering silence, or segmenting audio into speech chunks. Choose Silero VAD for short segments, SpeechBrain…

threejs

Three.js scene-graph parsing and export workflows: mesh baking, InstancedMesh expansion, part partitioning, per-link OBJ export, and URDF articulation.

pddl-skills

Automated Planning utilities for loading PDDL domains and problems, generating plans using classical planners, validating plans, and saving plan outputs. Supports standard PDDL parsing, plan synthesis, and correctness verification.

pymatgen

Materials science toolkit. Crystal structures (CIF, POSCAR), phase diagrams, band structure, DOS, Materials Project integration, format conversion, for computational materials science.

attitude-controller-planner

Use this skill when implementing the inner control loop for a quadrotor — attitude (roll/pitch/yaw) PID control and attitude planning (converting desired acceleration to desired Euler angles). Covers gain layout, integral reset pattern, and the attitude…

flight-plan-parser

Use this skill when converting natural language flight commands into waypoints and timing for a drone simulator. Covers parsing commands like "Take off to X m height in Y seconds", "Hover at X m height for Y seconds", "Fly from (x,y,z) to (x',y',z') in T…

motor-model-dynamics

Use this skill when simulating quadrotor physical dynamics — mapping desired thrust/moments to individual motor RPMs via a propeller allocation matrix, applying first-order motor lag, and integrating the nonlinear equations of motion (translational and…

plot-quadrotor

Use this skill when visualising drone simulation results. Produces three matplotlib figures — desired vs actual trajectories, instantaneous error, and cumulative absolute error — for all 5 state groups (position, orientation, velocity, angular velocity,…

position-controller-trajectory-planner

Use this skill when implementing the outer control loop for a quadrotor — position PID control (position/velocity error → thrust and desired acceleration) and trajectory planning from flight-plan waypoints (takeoff, hover, fly, land segments → smooth 15-row…

stepinfo-3d

Use this skill when computing 3D step-response performance metrics for point-to-point drone flight — rise time, settling time, percent overshoot, and steady-state error based on Euclidean distance to the final target. Use instead of 1D stepinfo for any flight…

transaction-concurrency-control-foundations

Provides database transaction concurrency-control background. Use when reasoning about transactions, commits and aborts, serializability, isolation anomalies, conflicts, dependencies, and protocol families such as locking, timestamp ordering, OCC, MVCC, SSI,…

transaction-protocol-reasoning

Teaches how to reason from a transaction protocol’s rules to a working model. Use when analyzing a concurrency-control paper, spec, or algorithm to identify state, metadata, invariants, operation rules, examples, counterexamples, guarantees, false aborts,…

transaction-trace-analysis

Teaches evidence-based analysis of transaction logs and traces. Use when reconstructing transaction behavior from tables, TSV/CSV files, timestamps, counters, object histories, validation records, abort records, retries, partial logs, or when producing…

ada-plan-view-accessibility

Use when checking simplified ADA-derived plan-view bathroom accessibility constraints such as turning space, door clear width, toilet centerline, grab bars, and lavatory knee/toe clearance.

architectural-dxf-extraction

Use when extracting plan-view architectural geometry from DXF files with semantic CAD layers, especially when outputs must normalize rooms, doors, fixtures, clearances, and grab bars into machine-checkable JSON.

geometric-layout-repair

Use when producing a minimally changed repaired 2D layout in JSON from extracted architectural geometry and rule-based clearance constraints.

skill-creator

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Codex's capabilities with specialized knowledge, workflows, or tool integrations.

mip-solver-and-solution-audit

Operational workflow for hard integer-programming optimization tasks: selecting an installed solver, preserving solver/incumbent certificates, extracting feasible schedules, recomputing metrics from final outputs, and writing consistent reports. Use when a…

ordered-window-sequencing-mip

Permutation and sequencing integer-programming formulations with ordered local-window variables, prefix/suffix continuity, and overlap penalties. Use when assigning exams, jobs, tasks, visits, blocks, or resources to ordered positions and costs depend on…

cache-policy-comparison

Compare and implement eviction policies (LRU, LFU, FIFO, S3FIFO, ARC) for bounded-capacity caches. Use when choosing or implementing an eviction policy for a buffer pool, page cache, CDN edge, or LLM KV cache, or when writing a replay simulator that supports…

prefix-cache-replay

Replay an LLM inference request trace (Mooncake / vLLM / SGLang hash_ids format) against a block-level KV prefix cache and compute hit statistics. Use when given a request trace plus cache configuration and asked for hit rate, hit tokens, or final cache…

grpo

Reference for the GRPO (Group Relative Policy Optimization) algorithm. Use when implementing, debugging, or verifying a GRPO training pipeline — covers the mathematical formulation (group-relative advantages, clipped surrogate loss, KL penalty), the training…

rl-post-training

Diagnostic guide for RL-based post-training of language models (GRPO, PPO, REINFORCE, DPO). Use proactively when debugging a training pipeline that shows no improvement, loss anomalies, reward stagnation, NaN gradients, or other unexpected behavior during…