ワンクリックでManusで任意のスキルを実行

bugmagnet

スター23

フォーク7

更新日2026年4月18日 21:01

Discover edge cases and test coverage gaps through systematic analysis. Use when analysing TEST coverage, hunting for edge cases, or when code-reviewer identifies test gaps. Triggers on "find holes in my tests", "what could go wrong", "is this well tested", "what edge cases am I missing". When to pick bugmagnet vs alternatives: - find-bugs → branch-diff bug scan (uses git diff, scoped to recent changes) - security-review → OWASP-categorised vulnerability audit - code-review → PR-style review with comment threads - ct:bugmagnet → test-gap edge-case discovery (THIS one) If the question is "are my TESTS missing cases?", use bugmagnet. Otherwise pick one of the three above.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

Manusで実行

ソース

pvillega

pvillega/claude-templates

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

BugMagnet

Systematic test coverage analysis and bug discovery.

Based on gojko/bugmagnet-ai-assistant.

Workflow

🔍 ANALYSE  → Understand code and existing tests
📊 GAP      → Identify missing coverage
✍️ WRITE    → Implement tests iteratively
🔬 ADVANCED → Deep edge case exploration (optional)
📋 SUMMARY  → Document findings and bugs

Interactive mode (default): STOP and wait for user confirmation between phases. Autonomous mode: When invoked by another agent (e.g. code-reviewer), skip all STOP points and proceed through all phases automatically. Present the full summary at the end instead of pausing for input.

Phase 1: Analysis (🔍 ANALYSE)

Detect language and testing conventions
Read implementation — public API, parameters, state, dependencies
Locate test file — if none exists, ask user about creating one
Run baseline coverage if tools available
Read existing tests — understand current coverage and patterns
Ask user: "Are there additional files I should review?"

STOP — Wait for user input. (Skip in autonomous mode.)

Phase 2: Gap Analysis (📊 GAP)

Evaluate missing coverage using edge-cases.md:

Boundary conditions, error paths, state transitions
Complex interactions, domain-specific edge cases
Violated domain constraints

Categorise: High (core, errors, boundaries) → Medium (interactions, state) → Low (rare, performance). Present analysis with specific examples.

STOP — Ask user which tests to implement. (Skip in autonomous mode — implement all high-priority tests.)

Phase 3: Test Implementation (✍️ WRITE)

For each test, highest priority first: write single test (or 2-3 related), name describes outcome ("returns X when Y"), run immediately. Maximum 3 attempts per test — document and move on.

→ While writing tests, actual bug discovered in implementation → STOP. Do NOT fix the bug.

Create minimal reproduction test.
Explore 2-3 adjacent code areas (bugs cluster).
Document in skipped test: brief description, root cause, code location, expected vs actual, proposed fix. → Return to test writing.

STOP — Ask user if they want advanced coverage. (Skip in autonomous mode — proceed to advanced if high-priority gaps remain.)

Phase 4: Advanced Coverage (🔬 ADVANCED)

Create separate test suite: "bugmagnet session <date>"

Work through edge-cases.md comprehensively — complex interactions, error handling, numeric/date/string/collection edge cases, state transitions, domain-specific cases.

Phase 5: Summary (📋 SUMMARY)

Keep summary under 30 lines. No preamble.

## Test Coverage Summary

**Tests Added:** X total (Y passing, Z skipped/bugs)

**Bugs Discovered:**
1. Bug name — file:line — root cause — proposed fix

Reference Files

Edge Case Checklist

このリポジトリの他の Skills

同じリポジトリ

fine-tuning-llms

pvillega/claude-templates

Deep LLM fine-tuning operational intuition — LoRA/QLoRA merge pitfalls, preference optimization (DPO/ORPO/SimPO/GRPO), chat-template footguns, completion-only loss masking, sequence packing, eval contamination, toolchain (TRL/axolotl/unsloth/torchtune). Load when post-training open-weights models, picking preference-optimization methods, configuring LoRA/QLoRA, suspecting eval contamination, or choosing a toolchain. Skip for inference/serving (use `llm-inference-serving`), prompt engineering, RAG, or pretraining. Triggers on: "QLoRA", "merge_and_unload", "DPO", "ORPO", "SimPO", "GRPO", "GRPOTrainer", "axolotl", "unsloth", "torchtune", "nf4", "apply_chat_template", "DataCollatorForCompletionOnlyLM", "cu_seqlens", "MMLU contamination", "catastrophic forgetting".

2026-05-0723

hetzner

pvillega/claude-templates

Deep Hetzner operational intuition — Cloud vs Robot vs Storage Box split, server-type taxonomy, region availability, IPv4 cost model, snapshot vs backup vs image, firewall default-deny, hcloud CLI, Floating-IP aliasing, cloud-init, Terraform provider drift. Load when sizing Hetzner resources, migrating to/from Hetzner, automating with hcloud/Terraform, or diagnosing Hetzner-specific gotchas. Skip for generic IaaS, Linux admin, or non-Hetzner clouds. Triggers on: "hcloud cli", "hcloud_server", "cax11", "cpx21", "fsn1", "hetzner robot", "primary ip cost", "without-ipv4", "vswitch", "floating ip hetzner", "storage box".

2026-05-0723

llm-inference-serving

pvillega/claude-templates

Deep LLM-inference-serving operational intuition — vLLM, SGLang RadixAttention, TensorRT-LLM tradeoffs; PagedAttention and KV-cache math, chunked prefill, prefix caching, disaggregated serving (DistServe, Mooncake), speculative decoding, quantization, MoE parallelism, TTFT/TPOT. Load when tuning serving-engine knobs, KV-cache sizing, batching design, quantization-format selection, speculative-decoding choice, or MoE layout. Skip for fine-tuning, embeddings serving, or hosted-API integration. Triggers on: "vLLM", "SGLang", "TensorRT-LLM", "PagedAttention", "RadixAttention", "chunked prefill", "prefix caching", "kv-cache-dtype", "EAGLE-3", "Medusa", "MLA", "AWQ", "NVFP4", "DistServe", "Mooncake", "EPLB", "TTFT", "TPOT", "goodput".

2026-05-0723

rag-ops

pvillega/claude-templates

Deep RAG application-layer operational intuition — chunking (late chunking, contextual retrieval), embedding-model landscape, reranker economics (ColBERT, ColPali), hybrid retrieval (BM25 + dense + RRF, SPLADE-v3), evaluation (RAGAS, judge bias), grounding, long-context-vs-RAG. Load when designing chunking, embedding-model selection, reranker integration, eval design, hallucination audits, or long-context-vs-RAG sizing. Skip for ANN index choice (use `vector-search`), embedding fine-tuning, or prompt engineering. Pairs with `vector-search`. Triggers on: "late chunking", "contextual retrieval", "MTEB", "BGE-M3", "Matryoshka embeddings", "SPLADE-v3", "RRF", "ColBERTv2", "ColPali", "HyDE", "RAGAS", "groundedness", "lost in the middle", "Self-RAG", "CRAG".

2026-05-0723

argocd

pvillega/claude-templates

Deep ArgoCD operational intuition — sync waves and hook lifecycle, ApplicationSet generators, Sync vs Health, drift root causes, sync options, AppProject constraints, rollback, multi-cluster. Load for sync ordering, hook design, drift diagnosis, generator selection, rollback planning, or multi-cluster registration. Skip for "what is GitOps", install walkthroughs, or basic Application YAML. Triggers on: "sync wave", "presync hook", "syncfail", "applicationset generator", "matrix generator", "hpa replicas drift", "ignoreDifferences", "ServerSideApply", "Replace=true", "self-heal", "app-of-apps cascade".

2026-05-0723

bash-hardening

pvillega/claude-templates

Deep bash correctness — `set -euo pipefail` traps, IFS/word-splitting, trap inheritance, process substitution, array gotchas, version-gated features, ShellCheck "looks right but isn't" patterns. Load when writing or hardening non-trivial bash scripts (installers, CI, unattended ops). Skip for basic if/while/for, one-shot interactive commands, or zsh/fish/POSIX-sh work. Triggers on: "set -e not triggering", "pipefail", "trap ERR", "PIPESTATUS", "macOS bash 3.2", "shellcheck SC2086", "harden bash script", "robust bash".

2026-05-0723

name	bugmagnet
description	Discover edge cases and test coverage gaps through systematic analysis. Use when analysing TEST coverage, hunting for edge cases, or when code-reviewer identifies test gaps. Triggers on "find holes in my tests", "what could go wrong", "is this well tested", "what edge cases am I missing". When to pick bugmagnet vs alternatives: - find-bugs → branch-diff bug scan (uses git diff, scoped to recent changes) - security-review → OWASP-categorised vulnerability audit - code-review → PR-style review with comment threads - ct:bugmagnet → test-gap edge-case discovery (THIS one) If the question is "are my TESTS missing cases?", use bugmagnet. Otherwise pick one of the three above.
credit	Adapted from channingwalton/dotfiles (https://github.com/channingwalton/dotfiles)

BugMagnet

Systematic test coverage analysis and bug discovery.

Based on gojko/bugmagnet-ai-assistant.

Workflow

🔍 ANALYSE  → Understand code and existing tests
📊 GAP      → Identify missing coverage
✍️ WRITE    → Implement tests iteratively
🔬 ADVANCED → Deep edge case exploration (optional)
📋 SUMMARY  → Document findings and bugs

Phase 1: Analysis (🔍 ANALYSE)

Detect language and testing conventions
Read implementation — public API, parameters, state, dependencies
Locate test file — if none exists, ask user about creating one
Run baseline coverage if tools available
Read existing tests — understand current coverage and patterns
Ask user: "Are there additional files I should review?"

STOP — Wait for user input. (Skip in autonomous mode.)

Phase 2: Gap Analysis (📊 GAP)

Evaluate missing coverage using edge-cases.md:

Boundary conditions, error paths, state transitions
Complex interactions, domain-specific edge cases
Violated domain constraints

Categorise: High (core, errors, boundaries) → Medium (interactions, state) → Low (rare, performance). Present analysis with specific examples.

STOP — Ask user which tests to implement. (Skip in autonomous mode — implement all high-priority tests.)

Phase 3: Test Implementation (✍️ WRITE)

For each test, highest priority first: write single test (or 2-3 related), name describes outcome ("returns X when Y"), run immediately. Maximum 3 attempts per test — document and move on.

→ While writing tests, actual bug discovered in implementation → STOP. Do NOT fix the bug.

Create minimal reproduction test.
Explore 2-3 adjacent code areas (bugs cluster).
Document in skipped test: brief description, root cause, code location, expected vs actual, proposed fix. → Return to test writing.

STOP — Ask user if they want advanced coverage. (Skip in autonomous mode — proceed to advanced if high-priority gaps remain.)

Phase 4: Advanced Coverage (🔬 ADVANCED)

Create separate test suite: "bugmagnet session <date>"

Work through edge-cases.md comprehensively — complex interactions, error handling, numeric/date/string/collection edge cases, state transitions, domain-specific cases.

Phase 5: Summary (📋 SUMMARY)

Keep summary under 30 lines. No preamble.

## Test Coverage Summary

**Tests Added:** X total (Y passing, Z skipped/bugs)

**Bugs Discovered:**
1. Bug name — file:line — root cause — proposed fix

Reference Files

Edge Case Checklist