Agent Skills by bianhaifeng789-hue
Showing 45 vetted skills indexed across 1 GitHub repositories.
harness-score-analyzer
Analyze QA score histories to categorize trends and recommend strategies.
harness-builder
Automates turning product specs and QA feedback into source files and commits changes.
harness-browser-test
Automate headless Chromium browser tests with scripted actions and screenshots.
parallel-tool-calls-integration
Enable concurrent tool call execution with ordered responses and error handling.
progress-persistence
Create structured checkpoints in progress.md with recent git diff context.
harness-planner
Expand 1-4 sentence briefs into a structured spec.md with required sections.
profile-selector
Match task keywords to profiles and return configuration with confidence and thresholds.
harness-evaluator
Automates QA scoring of web apps via Playwright tests and generates feedback.md reports.
profiles-system
Select runtime profiles using TB2 metadata for task execution.
text-only-nudge
Detect text-only plans and inject a nudge to force tool execution.
path-tracing
Reverse-engineer ray-traced images to recover camera, geometry, material, and lighting parameters.
video-processing
Automate video editing with FFmpeg for trimming, transcoding, and frame extraction.
install-windows-3.11
Install and run Windows 3.11 in QEMU with VNC and noVNC access.
chess-best-move
Detect chess pieces from board images and verify optimal legal moves.
reshard-c4-data
Reshard large file collections into hierarchical shards with per-directory limits.
gcode-to-text
Extract letters and symbols encoded in CNC, 3D-printer and laser-cutter G-code movement coordinates.
vulnerable-secret
Extracts secrets from vulnerable ELF binaries using static and dynamic analysis techniques.
compile-compcert
Build CompCert from source with the correct Coq/OCaml toolchain via opam.
train-fasttext
Train FastText supervised text classification models within accuracy and size constraints.
mailman
Configure Postfix and Mailman3 integration with LMTP delivery verification.
schemelike-metacircular-eval
Build and test metacircular evaluators that interpret themselves.
winning-avg-corewars
Develop CoreWars Redcode warriors to meet specified win rates via pmars battles.
llm-inference-batching-scheduler
Optimize LLM inference batching to minimize compilation overhead and padding waste.
extract-moves-from-video
Extract validated command sequences from video recordings using OCR.