780 件 ・ 関連度順
概要と使いどころ
Builds Stitch-ready prompts via two paths — Path A enhances vague ideas into polished prompts, Path B merges a Design Spec JSON + user request into a structured [Context] [Layout] [Components] prompt.
日本語の概要は準備中です。原文の説明を表示しています。
Explores codebases to find relevant files, trace execution paths, and map architecture with team communication capabilities for collaborative analysis (converted from agent)
日本語の概要は準備中です。原文の説明を表示しています。
Reference patterns for graph algorithm problems including BFS, DFS, Dijkstra, topological sort, union-find, MST, Bellman-Ford, and bipartite checking. Provides recognition signals, Python templates, edge cases, and common mistakes for each technique. Use when solving problems involving graphs, trees, shortest paths, connectivity, or dependency ordering.
日本語の概要は準備中です。原文の説明を表示しています。
Use when working on imp's quantization formats, loaders, or dequant paths - GGUF Q4_0...Q8_0/Q*_K/IQ4/MXFP4, NVFP4 two-level scaling (Modelopt vs compressed-tensors layouts), FP8 E4M3, StorageTier, decode cache, KV-cache dtypes (auto/fp16/fp8/int8/int4/nvfp4/mxfp4), "which quant should I use", "which KV dtype", scale-factor layout, dequant kernel wiring, imp-quantize / AWQ, judging quant quality (PPL). Do NOT use for writing/optimizing GEMM/GEMV kernels (sm120-cuda-expert) or measuring quant perf (benchmark-cuda).
日本語の概要は準備中です。原文の説明を表示しています。
Use when asking whether something in imp is actually finished - hunting stubs, placeholders, request fields that are parsed and then ignored, code paths or kernels that never run, gate-based features that silently no-op, tests that assert nothing. Triggers on "is this implemented", "unfinished", "stub", "placeholder", "dead path", "does this flag do anything", "accepted but ignored", "kernel never launched", "test asserts nothing", "neutral A/B on a gated feature". Do NOT use for structural debt / god files / duplication (codebase-audit) or for who-calls-what (code-graph).
日本語の概要は準備中です。原文の説明を表示しています。
Use when imp outputs differ between two paths that should agree - chunk sizes, prefill vs decode, prefix-cache resend, two engines (imp vs llama.cpp vs HF), two builds or kernels - "output changes with chunk size", "first token differs", "logprobs moved", "which layer diverges", "who is right, imp or llama.cpp", "near-tie". Do NOT use for repetition loops or garbage output (check-degeneration), a new arch that loads wrong (add-model-arch), or perf (benchmark-cuda).
日本語の概要は準備中です。原文の説明を表示しています。
Living catalogue of 7 known hazard entries (SE-01 through SE-07) specific to agent-studio: Windows backslash paths, prototype pollution, hook exit codes, async swallowing, ReDoS in glob-to-regex, DST arithmetic, array mutation during iteration. Each entry: symptom, root cause, fix, test assertion.
日本語の概要は準備中です。原文の説明を表示しています。
Python dependency management with Poetry and Rye -- lockfile-driven workflows, dependency groups, monorepo patterns, and migration paths. Covers pyproject.toml-centric packaging for projects not yet on uv.
日本語の概要は準備中です。原文の説明を表示しています。
Decision matrix for selecting multi-agent topologies (Supervisor, Swarm, Hierarchical, Conductor) with token economics, failure modes, and escalation paths
日本語の概要は準備中です。原文の説明を表示しています。
Structured workflow for upgrading Next.js applications across major versions. Use when migrating a Next.js project from one major version to another (e.g., 13 to 14, 14 to 15, 15 to 16). Covers codemod automation, breaking change detection, incremental migration paths, and post-upgrade validation.
日本語の概要は準備中です。原文の説明を表示しています。
Create structured incident response runbooks with step-by-step procedures, escalation paths, and recovery actions. Use when building runbooks, responding to incidents, or establishing incident response procedures.
日本語の概要は準備中です。原文の説明を表示しています。
Deeply analyzes Agent Studio framework structural health: catching phantom require() references, wrong module depth paths, missing skill/agent dependencies, bloated configurations, archived references in active code, stale catalog counts, and empty tool/skill directories.
日本語の概要は準備中です。原文の説明を表示しています。
Single gateway for all core↔tech-pack interactions. Reads manifests, resolves paths, loads skills/agents, routes commands.
日本語の概要は準備中です。原文の説明を表示しています。
Voice a papercut reel and time it to the picture. Three paths, local Qwen3-TTS, your own recording or another TTS, or a placeholder voice. Covers delivery rules, voice.json, captions and scene timing. Use for "set up the voice", "re-voice this", "time the script".
日本語の概要は準備中です。原文の説明を表示しています。
Build a team Docker image from a manifest in .archon/teams/. Resolves plugin paths, generates Dockerfile with additive copy-only, makes plugins read-only, bakes team CLAUDE.md.
日本語の概要は準備中です。原文の説明を表示しています。
Map user journeys, onboarding paths, entry points, first actions, trust points, accessibility risks, and SEO/LLM handoffs for public websites without redesigning the visual system. Use this skill when the user asks how new readers, returning users, project reviewers, community members, buyers, or AI agents should move through a site and what each page must make clear before layout implementation.
日本語の概要は準備中です。原文の説明を表示しています。
Redact a network/MITM capture directory — JSONL, JSON, logs, HAR-style dumps — in place before it can be referenced publicly, scrubbing credential tokens, API keys, service ids, UUIDs, emails, home paths, usernames, and device or session hashes while preserving token-class prefixes for analysis and leaving public identifiers untouched. Covers class-preserving substitution, an idempotent re-runnable pass, distinguishing a secret from a public id, and verifying the scrubbed capture through the redaction gate. Use when a wire capture from a sanctioned probe must enter notes, a guide, or a public repo without leaking the session that produced it.
日本語の概要は準備中です。原文の説明を表示しています。
Design a generative recipe DSL so agents emit small validated JSON documents and ONE audited interpreter renders them all — data, not code. Covers the N-artifacts threshold for choosing a DSL over bespoke code, designing roles as style slots so a single recipe renders unchanged in every output style, the validator gate (compile + bounds + real render in all styles, plus a --fast mode for corpus-wide sweeps), hand-authored seed recipes, and the emergent conventions that keep a corpus of hundreds consistent without per-item steering. Use when many similar artifacts (icons, status badges, diagram nodes, test fixtures) must be generated by agents, when per-item bespoke code is multiplying unreviewable code paths, or when a whole corpus must be re-rendered cheaply into new styles.
日本語の概要は準備中です。原文の説明を表示しています。
Remove dead code, unused imports, fix lint warnings, and normalize formatting across a codebase without changing business logic or architecture. Use when lint warnings have piled up during rapid development, unused imports and variables clutter files, dead code paths were never removed, formatting is inconsistent, or static analysis tools report fixable hygiene issues.
日本語の概要は準備中です。原文の説明を表示しています。
Modern Python idioms for any Python 3.10+ project: PEP 8 style, PEP 484/526 type hints, dataclasses, pathlib, enums, f-strings, structural pattern matching (match/case), context managers, exception handling, and null discipline. Stack-agnostic — referenced by every Python plugin in the marketplace. Use this skill to: - Write strongly-typed code with type hints on every function signature and class attribute. - Replace isinstance chains and magic constants with match/case and enums. - Use dataclasses or NamedTuple for value objects and data containers. - Handle resources safely with context managers and pathlib for all file paths. - Apply PEP 8 naming and structure conventions consistently. Do NOT use this skill for: - Framework-specific idioms (Django ORM, FastAPI routers, Flask blueprints — those live in framework plugin skills). - Packaging and dependency management — see python-foundation:python-tooling. - Testing patterns — see python-foundation:pytest-testing.
日本語の概要は準備中です。原文の説明を表示しています。
Use when Rust code, especially unsafe or panic-critical paths, needs a Kani, Verus, or Creusot harness written, run, and its failure read. Not for choosing the proof policy: use proof-driven.
日本語の概要は準備中です。原文の説明を表示しています。
Diagnose React + Three.js (or any browser canvas) frame stutter by capturing a CDP CPU profile via Playwright instead of guessing from `[perf]` console logs. Use when: (1) frame time is bad but `[perf]` instrumentation doesn't pinpoint the offender, (2) you've spent more than ~30 minutes "fixing" suspects without measuring, (3) hot paths involve animation systems, allocation bursts, or invisible scene-graph traversal, (4) you need self-time per function with line numbers, not aggregate "render slow". The `console.warn` pattern is good for monitoring; CDP is needed for diagnosis. Covers script template, sourcemap setup so minified function names are decoded, and interactive variant where the user drives the interaction and signals stop via `touch /tmp/<flag>`.
日本語の概要は準備中です。原文の説明を表示しています。
Recover from stale RFDB nodes that survive `commit_batch` cleanup despite having their `file` field listed in `changed_files`. Use when: (1) re-running an enricher/orchestrator step but old data with the same file field stays in the graph, (2) `commit_batch(changed_files=[X], ...)` reports success but a query later still returns nodes with `file = X`, (3) you re-pointed structural nodes (e.g., DIRECTORY/FILE) from a synthetic file path to real file paths and the old synthetic-path nodes won't go away, (4) graph-stream / find_by_type returns a mix of "stale" and "new" nodes for the same logical entity. Root cause: `handle_commit_batch` calls `engine.find_by_attr({file: X})` to enumerate nodes-to-delete, but that query path returns only a PARTIAL set after compaction (L1 segments and the file→nodes index can desync). Workaround: enumerate stale nodes via the TS rfdb client `queryNodes({file: X})` (different code path that finds all of them) and call `deleteNode(id)` per node, then re-commit the intended state. Related: rfdb-manifest-l1-carryforward (same L1 family).
日本語の概要は準備中です。原文の説明を表示しています。
Use when the user wants to run SQL - especially analytical SQL - on local files (parquet/csv/json), URLs, S3 paths, or remote databases (Postgres, MySQL, MongoDB, ClickHouse Cloud, Iceberg, Delta Lake) without setting up a server. Provides chDB - embedded ClickHouse SQL in Python with 1000+ functions, Session for stateful multi-step pipelines, parametrized queries, and cross-source joins via `s3()`, `mysql()`, `postgresql()`, `iceberg()`, `deltaLake()`, `remoteSecure()` table functions. TRIGGER when: user wants SQL on parquet/csv/files or across remote analytical sources; uses ClickHouse SQL features (window functions, windowFunnel, geoToH3, JSON path ops, Session, parametrized queries); imports `chdb` or calls `chdb.query()`. SKIP this skill for pandas-style DataFrame method-chaining (use chdb-datastore instead) or ClickHouse server administration.
日本語の概要は準備中です。原文の説明を表示しています。