AI Agent Skills

Maintain bounded, durable AI project memory in a repository's docs/ai/ pack. Use when Codex or Claude Code needs to create, read, update, or audit project memory (projec…

Query and maintain the user's local LLM Wiki from any Codex or Claude Code conversation or project. Use when the user asks to search personal knowledge, recall prior res…

在风险分析完成后、写测试代码之前，生成中文业务用例并等待用户确认。这是整个 QA 流程中唯一的强制人工门禁——确认后不可回头改用例。 从 context、risk-analysis、已有用例索引出发，生成只含业务行为的 test-cases.json（不含技术检查如 Maven/build/安装）， 渲染 test-cases.html…

Compare two versions of an LLM-directed document - an original (teacher) and a candidate (student) - across a transfer set and return a per-case behavioural-equivalence…

Attacker's-eye test of LLM/agent defenses: instruction hijacking, data exfiltration and tool abuse through untrusted content; verifies whether the defense actually holds.

Build AI-powered software with TypeSafe: small units of AI intelligence you can use like programming primitives. Its System One models, including Jev, turn natural langu…

Use when running or configuring `xmake ai` — the terminal coding agent provided by the xmake-harness addon. Covers installing it, setting an llm provider and api key, se…

Query integrated indexes using text with Pinecone MCP. IMPORTANT - This skill ONLY works with integrated indexes (indexes with built-in Pinecone embedding models like mu…

Run an arbitrary Hugging Face safetensors model on an Intel GPU using **upstream PyTorch** (>= 2.8) with the built-in `torch.xpu` device. Covers loading from the Hub, pi…

Abuse an LLM agent's tools/functions — coerce it to call tools with attacker-chosen args for SSRF, RCE, data exfil, or privilege abuse. Load when the target is an agent…

Bypass an LLM's safety/guardrails to make it produce restricted output or ignore its policy. Load when testing an AI product's content controls, "jailbreak", "guardrail…

Unbounded-consumption / denial-of-wallet attacks on LLM apps — force runaway tokens, cost, or latency. Load when testing an LLM product's limits/billing, on "LLM DoS", c…

Extract or steal an ML/LLM model's parameters, training data, or system prompt via query access — model stealing, membership inference, training-data extraction. Load wh…

Use when designing, debugging, training, researching, or evaluating an AI agent and the decision concerns one of seven areas: minimum agent loop, inference-time compute…