发现 Skills

Evaluate session interaction logs and provide actionable feedback on user-AI collaboration. Use when asked to "review session", "evaluate logs", "give feedback", "analyze interactions", "session feedback", or "improve workflow". Reads logs from logs/copilot/ and generates a feedback report under logs/feedback/report-{date}.md with strengths, improvement areas, and concrete suggestions for future sessions.

Fast binary analysis with string reconnaissance and static disassembly (RE Levels 1–2). Use when triaging suspicious binaries, extracting IOCs quickly, or performing initial malware analysis. Completes in ≤2 hours with automated decision gates.

Work on novel-specific features including plot engines, character development, world-building, timeline management, and GOAP-based story generation. Use when implementing narrative systems, character arcs, or story planning tools.

The patch looks good and addresses the identified issues. Let me verify the skill file exists and then output the complete evolution.

Extract PDF text content using shell tools or Python libraries when the read_file PDF handler fails

SQL analysis skill for Ascend PyTorch Profiler / msprof DBs (e.g., ascend_pytorch_profiler*.db, msprof_*.db). Converts natural-language questions (operator time, communication, dispatching, scheduling, schema/table queries) into safe, executable SQL and, when needed, extracts table-structure details from official documentation.

Tool to prepare Diffusers model weights for Huawei Ascend NPU. Supports downloading model weights from HuggingFace and ModelScope, and generating synthetic placeholder weights for validation based on model_index.json and each component's config.json. Intended for use when users need to download Diffusers model weights or generate test weights.

OligoFormer Ascend NPU migration Skill for migrating PyTorch Transformer–based siRNA efficacy prediction models to Huawei Ascend NPU. Covers environment setup, RNA-FM dependency installation, code adaptation, inference validation, and optional training workflows.

Specialized agent for creating and improving Connection Properties in Aspire resource and README files

Deep knowledge of GitHub Actions, Railway, Supabase, and Postgres platforms. Use when troubleshooting, configuring, or optimizing any of these platforms.

Get an external senior game developer's critique of the project. Reviews usefulness, identifies issues, and suggests features.

Guide for creating effective skills. Use this skill when you want to create a new Skill (or update an existing one) that extends Claude with specialized knowledge, workflows, or tool integrations.

Design scalable, reliable software systems. Use when planning new systems, major features, or architecture changes. Covers C4 diagrams, trade-off analysis, and system decomposition.

Use when you're stuck on complex problems, need architecture validation, want code reviews from multiple perspectives, or debugging hits a wall — guides leveraging multiple AI models in parallel for better results.

Use when a plan must be stress-tested against the codebase's existing language and decisions — grill it against current docs, glossary terms, and ADRs before implementation begins

Use when you need an independent second opinion before, during, or after implementation — run challenge, consult, or review mode in a direct builder-to-builder voice

Use when creating or developing, before writing code or implementation plans - refines rough ideas into fully-formed designs through collaborative questioning, alternative exploration, and incremental validation. Don't use during clear 'mechanical' processes

Tamper-resistant audit watchdog for Clawdbot agents. Detects and logs suspicious filesystem activity with HMAC-chained evidence.

Use when you have a written implementation plan to execute in a separate session with review checkpoints

Create JIRA tickets using the jira-create CLI tool with sensible defaults for the Program Architecture team

Use when orchestrating multi-step animations — page transitions, onboarding flows, wizard steps, complex reveals, or any choreographed animation sequence.

Kubernetes cluster, pod, node, and workload monitoring. Use when analyzing K8s health, resource optimization, pod failures, OOMKills, scheduling, or security posture. Also use for Kubernetes operational events like pod restarts, OOM events, evictions, and cluster event history. Trigger: "Kubernetes pods", "K8s cluster health", "OOMKill", "pod restarts", "container CPU", "namespace resource usage", "over-provisioned pods", "privileged containers", "pod placement", "K8s node capacity", "running containers by cluster", "workload scheduling", "pod evictions", "K8s labels and annotations", "kubernetes events", "pod restart events", "OOM events", "K8s event history". Do NOT use for explaining existing queries, product documentation questions, AWS-specific resource queries, service-level RED metrics, distributed tracing, or log analysis — use the relevant skill instead.

Manage Dynatrace environments using dtctl - install/update the CLI, configure authentication with OS keyring tokens, and run kubectl-style operations (get/describe/edit/apply/delete/query/exec) for workflows, dashboards, notebooks, DQL, SLOs, settings, buckets, lookups, OpenPipeline, and Davis AI. Use when the user wants to control Dynatrace resources via dtctl.

Import ADO pipelines for enabled AI automation agents and set all required pipeline variables. Profile-aware — imports only pipelines for the configured profile (consumer or full-hub). Reads config from .ai/automation/infra.json.