d-experts
Official@d-experts
d-experts provides a complete ICFP Contest 2026 Littleman toolkit for designing, judging, optimizing, and visualizing compact .man programs at team scale.
Agent Skills by d-experts
Showing 12 vetted skills indexed across 1 GitHub repositories.
d-expertslittleman-compaction-idioms
Documents proven idioms for reducing footprint and ticks in ICFPC 2026 Littleman programs.
d-expertsplace-littleman-rooms
Places and routes Littleman rooms into compact verified .man programs.
d-expertsicfpc-local-judge
Judge ICFPC Littleman solutions with the official WASM engine and manage submissions.
d-expertsicfpc-module-synthesizer
Synthesizes instruction-minimal Littleman modules from finite behavior tables using exact BFS and CEGIS.
d-expertsicfpc-layout-optimizer
Compresses ICFPC 2026 Littleman .man layouts via simulated-annealing placement and exact-length pipe routing.
d-expertsicfpc-problem-notes
Reads and updates per-problem knowledge notes for ICFPC 2026 Littleman contest work.
d-expertsicfpc-trace-optimizer
Profiles Littleman .man programs via official WASM per-tick traces to locate execution hotspots and benchmark local rewrites.
d-expertsicfpc-aws-compute
Runs CPU-heavy ICFPC explorations as AWS Batch Spot array jobs and downloads results.
d-expertsicfpc-session-archive
Scan, sanitize, verify, and upload Codex or Claude Code session exports to S3.
d-expertsvisualize-littleman-run
Render Littleman .man executions as SVG/PNG snapshots with movement and stall heatmaps.
d-expertsicfpc-littleman
Solves ICFPC 2026 littleman problems by writing, testing, and optimizing .man programs with the lman simulator.
d-expertsoptimize-littleman-solutions
Audits and optimizes existing ICFPC littleman .man solutions to reduce footprint, ticks, and score.
Frequently Asked Questions About d-experts
FAQPage SchemaWhat tasks can I accomplish with d-experts' ICFPC Littleman skills?▼
You can solve Littleman problems, judge and score .man solutions locally with the official WASM engine, compress layouts via simulated annealing, synthesize minimal modules from behavior tables, profile per-tick traces, render execution heatmaps, and manage best-solution history and official team submissions.
Who are these Littleman skills designed for?▼
They target competitive programming teams competing in ICFP Contest 2026, especially members optimizing .man footprint and tick scores, running parallel cloud searches, maintaining per-problem knowledge notes, and archiving Codex or Claude Code session exports for team review.
How do I judge and submit a Littleman solution?▼
Use the icfpc-local-judge skill: it runs the official WASM engine locally, scores a .man file, maintains a content-addressed best-solution store, and queues official submissions through the team proxy. It resolves contest IDs like 1-1, API slugs, or problem names, with a live localhost HTML viewer.
What compute resources are needed for large searches?▼
Heavy exploration uses the icfpc-aws-compute skill on the repository's AWS Batch EC2 Spot environment. Independent seeds are sharded as array jobs, with monitoring, cancellation, and per-index artifact download. Local judging, layout optimization, and synthesis run on a single machine.
What are the prerequisites for using these skills?▼
You need the ICFPC 2026 Littleman repository with its local simulator lman, the official WASM judge engine, and problems/<slug>/ folders. Cloud search requires configured AWS Batch EC2 Spot access; session archiving uses S3 with the team proxy's submittedBy identity.