AI & ML

aibot88 Bundle

KDL config reference for slop-mcp scopes, manage_mcps params, auth, metadata. 配置格式、範圍、認證、元數據之參考。 Use when: registering servers, inspecting auth, understanding KDL format, checking scope behavior.

aibot88 Bundle

Collection of expert-level TRAE agent system prompts for code generation, debugging, optimization, security auditing, and workflow acceleration. Reference library — use prompts as templates for specialized agent behavior.

aibot88 Bundle

Use BEFORE shipping high-stakes output, considering an architectural alternative, working with compliance/legal/safety-critical content, or stuck in a debug loop >2 cycles — fork the question to an independent source of truth (different model, fresh subagent, docs) before trusting your first answer. The meta-decision skill that fires existing review skills (second-opinion, adversarial-reviewer, dispatching-parallel-agents) at the right moments. Anchoring on first instinct is a known LLM failure mode; the cost of a 30-second outside review is far lower than the cost of shipping the wrong pattern.

aibot88 Bundle

Multi-agent parallel code review system inspired by Claude Code's /ultrareview. Spawn 5-20 specialized agents (security, logic, performance, edge cases, architecture) to review code simultaneously, cross-validate findings, and produce a consolidated bug report. Use when asked to do deep code review, ultrareview, multi-agent review, comprehensive bug hunting, security audit, or thorough pre-merge code inspection.

aibot88 Bundle

Draft, validate, split into threads, open X/Twitter composer links, and use the dry-run-first MCP workflow for live X posting when local credentials are explicitly configured. Use when Codex needs to prepare social launch copy, repo/project announcements, X posts, tweet threads, or shareable publishing drafts where the user should manually review before posting.

aibot88 Bundle

Analyze changes and create a meaningful commit with agent authorship. Internal skill for evolve_loop.

aibot88 Bundle

Give the AI agent its own EVM wallet with admin-controlled policies the agent CANNOT bypass even under prompt injection. Encrypted keystore (AES-256-GCM, scrypt KDF), policy file the agent has no tool to write, deterministic policy gate on every signing operation, optional local HTTP dashboard. Triggers: agent wallet, give the agent a wallet, agent address, fund the agent, agent autonomy, policy gate, kill switch, agent permissions, bounded autonomy, ERC-4337 alternative, session-key alternative.

aibot88 Bundle

Use when adding capabilities to an existing agent project — memory, app integration, VPC, multi-agent, migration, model changes, browser, code interpreter, or resource removal. Triggers on: "add memory", "remember across sessions", "call agent from app", "invoke agent from code", "auth to call agent", "streaming responses", "VPC", "VPC connectivity", "VPC error", "can't reach from VPC", "multi-agent", "A2A", "A2A auth", "orchestrator not delegating", "specialist not called", "migrate Bedrock Agent", "after import", "migration issue", "framework for migration", "change model", "browser tool", "code interpreter", "delete agent", "tear down", "agentcore remove", "cross-account memory", "resource-based policy on memory". Not for connecting to external APIs via Gateway — use agents-connect. Not for scaffolding a new project — use agents-get-started. Not for CLI/dev server errors — use agents-debug. Strands vs LangGraph in a migration context routes here.

aibot88 Bundle

Reference for the `.ato` declarative DSL: type system, connection semantics, constraint model, and standard library. Use when authoring or reviewing `.ato` code.

aibot88 Bundle

Author a new Cody skill and ship it end-to-end. Writes skills/<name>/SKILL.md, commits on the secret-agent-skills-bank branch (Cody's own branch), and fast-forwards main to the same commit so the existing publish-gateway-bundle workflow rebundles skills/** and pulls them onto the gateway. No human review step. Use when the user asks Cody to create a new skill. Do NOT use to edit an existing skill.

aibot88 Bundle

Discover, message, and coordinate AI coding agents (Claude Code, Codex CLI) running on the same machine via a shared Unix socket broker backed by SQLite. Supports cross-session task delegation, peer summaries, and session lifecycle management. Triggers on 'peer agents', 'cross-session communication', 'message Codex', 'inter-agent messaging', 'coordinate agents'.

aibot88 Bundle

Use this skill whenever the user mentions API keys, passwords, tokens, credentials, secrets, .env files, OAuth tokens, or any sensitive value that needs to end up in code, config, or a command. The skill stores secrets in a vault that the model never reads — you reference them by name and let the claude-vault CLI inject the actual values.

aibot88 Bundle

Create and manage TMDD threat models grounded in actual codebase architecture. Use when the user wants to threat-model a system, add a feature, create security threat mappings, run tmdd commands, or work with .tmdd/ YAML files.

aibot88 Bundle

Internal skill used by dev-implement during Phase 5 of /dev workflow. NOT user-facing - should only be invoked by dev-ralph-loop inside each implementation iteration. Handles Task agent spawning with TDD enforcement and two-stage review (spec compliance + code quality).

aibot88 Bundle

One-call structured triage of a git diff. Returns per-file role classification (source/test/config/doc/generated/build/fixture/migration), risk tier (low/medium/high) with reasons, public-API touch detection, co-changed-test detection, secret-risk path flagging, and aggregate stats. Replaces the 5-20 Read calls an agent burns reading each modified file individually to figure out "what kind of change is this and how risky is it". Stateless, no network, stdlib Python only.

aibot88 Bundle

Extracts JIRA issue keys from PR context and enriches with ticket details using claude.ai Atlassian MCP. NON-BLOCKING. Also scans fetched content for prompt injection.

aibot88 Bundle

Helpt bij het implementeren van LLM-specifieke beveiligingscontrols voor overheidstoepassingen, gebaseerd op de OWASP LLM Top 10, BIO2, NIS2 en AVG. Biedt prompt injection detectie, output sanitization, Data Loss Prevention (DLP), PII-filtering, data-classificatie voor LLM-input en beveiligingstests. Gebruik deze skill wanneer de gebruiker vraagt over 'LLM security', 'LLM beveiliging', 'AI beveiliging', 'AI security', 'prompt injection', 'prompt injectie', 'jailbreak', 'jailbreak detectie', 'OWASP LLM', 'OWASP LLM Top 10', 'output sanitization', 'output sanering', 'PII filtering', 'PII detectie', 'data loss prevention LLM', 'DLP LLM', 'data classificatie AI', 'data classification AI', 'LLM input guard', 'input validatie LLM', 'sensitive data exposure LLM', 'excessive agency', 'model denial of service', 'supply chain LLM', 'training data poisoning', 'insecure output handling', 'output handling', 'LLM red teaming', 'AI red team', 'LLM pentest', 'AI pentest', of wanneer de gebruiker een LLM-applicatie wil beveil

aibot88 Bundle

Use when working with the MCP server endpoint, JSON-RPC handling, tool registry, authentication, or files in packages/mcp/

aibot88 Bundle

Use when the user says "/bugcheck", "/mxBugChecker", "check for bugs", "find bugs", "audit for vulnerabilities", "verify the code", "look for issues in this file", or otherwise requests bug analysis on VCS changes or specific files. Verified-knowledge bug finder — every finding requires concrete code proof. Analyzes logic errors, runtime issues, edge cases, error handling, concurrency, resource leaks, security vulnerabilities, and performance regressions. Loads project context from the mxLore Knowledge-DB via MCP and persists findings via Skill Evolution.

aibot88 Bundle

(Industry standard: Routing Agent / Orchestrator Pattern) Primary Use Case: Analyzing an ambiguous trigger and routing it to one of the specific specialized implementations. Routes triggers to the appropriate agent-loop pattern. Use when: assessing a task, research need, or work assignment and deciding whether to run a simple learning loop, red team review, dual-loop delegation, or parallel swarm. Manages shared closure (seal, persist, retrospective, self-improvement).

aibot88 Bundle

OSINT Investigator v2.1 — comprehensive open-source intelligence skill. Triggers on: OSINT, recon, digital footprint, dorking, social media investigation, username lookups, email tracing, domain recon, entity mapping, OPSEC, image verification, metadata analysis, threat intel, people search, background research. Slash commands: /dork, /recon, /pivot, /entity, /timeline, /analyze-metadata, /verif-photo, /sock-opsec, /report, /simple-report, /full, /track, /link, /entities, /confidence, /export-entities, /import-entities, /compare, /timeline-entity, /find-path, /visualize, /stats, /export-graph, /risk-score, /anomaly, /pattern, /threat-model, /sanitize, /export-risk, /wizard, /template, /simple-mode, /progress, /save-checkpoint, /load-checkpoint, /qa-check, /coverage, /gaps, /verify-sources. Professional playbooks: journalist verification, HR background checks, cyber threat intel, private investigation. Integrations: Maltego, Obsidian, Notion.

aibot88 Bundle

Post-recording cleanup for minutes — wait for processing, update title from calendar, identify speakers, re-ingest knowledge base, add action items to reminders. Privacy-sensitive — use a local model only.

aibot88 Bundle

All-in-one prompt engineering competition toolkit — attack generation, defense hardening, real-time analysis, and pattern reference for AI security tournaments like Clash of Prompts. Use when preparing for or competing in prompt engineering competitions, red-teaming LLMs, hardening system prompts, analyzing prompt injection techniques, or practicing timed prompt challenges. Trigger phrases include "prompt clash", "prompt competition", "prompt tournament", "red team this prompt", "harden this prompt", "attack this prompt", "defend this prompt", "prompt injection", "prompt security", "clash of prompts", "prompt challenge", "prompt battle".

aibot88 Bundle

Write and refine production-grade prompts and tool definitions for Claude. Use when the user needs to craft a system prompt, improve an existing prompt, design tool schemas (ACI), or apply techniques like few-shot, XML structuring, role prompting, or chain-of-thought. Covers Claude 4.6 specifics. Do NOT use for skill authoring (use skill-author) or agent architecture (use agent-patterns).

aibot88 Bundle

Active-reading scaffolding for nonfiction books (EPUB). Parses an EPUB, generates a structured book analysis in the user's Obsidian vault at Read/{Book Name} ({Author})/, with 0. General.md as the book-level overview and one numbered markdown file per chapter containing pre-read brief, the user's own notes section, and post-read synthesis — all through configurable role lenses (e.g. PM, CEO, Entrepreneur). Six sub-skills — `reading-lens:setup` loads an EPUB with lens selection, `reading-lens:brief` generates a pre-read brief for a chapter, `reading-lens:synthesize` generates a post-read synthesis, `reading-lens:overview-redo` regenerates 0. General.md, `reading-lens:lenses` manages global lens presets, `reading-lens:to-notion` mirrors the book to Notion via MCP. Use whenever the user wants deep, structured book analysis, says "analyze this book", "read this book", "reading lens", "book analysis", "pre-read brief", "chapter synthesis", or references an EPUB path. Prefer this skill for any book-study workflow —

aibot88 Bundle

Amical platform help — open-source, local-first AI dictation app with Whisper STT, context-aware formatting, Ollama/OpenRouter LLM, MCP voice commands, and 100+ language support (MIT license). Use when setting up Amical on macOS or Windows for the first time, choosing between Whisper model sizes for accuracy vs speed, configuring Ollama for fully offline text formatting, context-aware formatting not adapting to the active application, microphone not switching or audio input issues, custom vocabulary not recognizing industry terminology, comparing Amical to Wispr Flow or Superwhisper for privacy-first dictation, or setting up voice commands with MCP integrations. Do NOT use for picking between meeting note-takers generally (use /sales-note-taker) or reviewing a single call for coaching (use /sales-call-review).

aibot88 Bundle

Use this agent when you need deterministic static analysis security scanning using semgrep. This agent complements security-sentinel by running rule-based pattern matching to catch known vulnerability signatures, hardcoded secrets, insecure function calls, and CWE patterns that LLM-based review may miss. Requires the semgrep CLI to be installed.

aibot88 Bundle

Create and refine high-quality Agent Skills (SKILL.md + bundled resources). Use when the user wants to create a new skill, improve an existing one, or turn a workflow into a reusable skill package. Covers structure, progressive disclosure, conciseness, feedback loops, and the quality checklist. Do NOT use for skill validation (use skill-check) or for prompt engineering (use prompt-craft).

aibot88 Bundle

Security audit for AI agent skills before installation. Scans SKILL.md files, hooks, scripts, and MCP configs for prompt injection, data exfiltration, credential theft, and malicious automation patterns. Use when installing a new skill, reviewing a skills repo, or auditing existing installed skills. Trigger phrases: "verify this skill", "is this skill safe", "audit skill", "check before installing", "scan this plugin".

aibot88 Bundle

Scaffold slop-mcp monitor session — verify PATH binary, choose generic event stream / SLOP polling script / timed session, emit copy-ready Claude Code Monitor({...}) JSON. 搭建 slop-mcp 監視會話。 Use when: starting Claude Code Monitor over slop-mcp, watching git/build/CI events, polling MCP for delta changes, running bounded watch session.

aibot88 Bundle

Generate a threat model from spec.md using STRIDE methodology. Use when you need to identify security threats, attack surfaces, and mitigations for a feature before implementation. Triggers on "threat model", "security analysis", "attack surface", "STRIDE analysis".

aibot88 Bundle

Use this skill when the task involves creating, reviewing, or refactoring installable skills, shared skill references, or reusable agent-authoring guidance; when the work depends on routing-quality descriptions, non-use boundaries, naming, progressive disclosure, or cross-skill authoring conventions; or when you want to normalize skill and guidance structure without turning the task into a full customization-builder workflow. Do not use it for end-to-end customization generation, normal code implementation, or general repository planning.

aibot88 Bundle

This skill should be used when the user asks about "Zod security", "Zod over-posting attack", "mass assignment Zod", "z.strictObject security", "z.custom security", "Zod coercion vulnerability", "Zod default PATCH bug", "Zod data loss", "Zod reportInput PII", "exposing ZodError to client", "Zod security review", "reviewing Zod validation code", or when performing a code review on TypeScript files that use Zod. Provides a comprehensive security model and review checklist for Zod usage.

personamanagmentlayer Bundle

Build reliable applications on large language models: prompt design, structured output, evaluation, guardrails, and cost and latency control. Use when the user mentions LLMs, prompts, prompt engineering, few-shot examples, structured or JSON output, function calling, hallucination, model evaluation, token costs, streaming, or when the task involves choosing a model, writing a system prompt, or making model output dependable enough for production.

personamanagmentlayer Bundle

Expert in Dynamics 365 platform, model-driven apps, canvas apps, Power Automate flows, Dataverse, plugins, and custom workflows. Use when the user mentions Microsoft platforms, ERP, CRM, business apps, the Power Platform, or Dataverse, or when the task involves Dynamics 365 Applications, Dataverse Platform, Customization & Development, or Plugin Development.

personamanagmentlayer Bundle

Build LLM agents that use tools safely: tool design, the agent loop, memory, MCP servers, multi-agent orchestration, sandboxing and prompt-injection defence. Use when the user mentions AI agents, tool use or function calling, MCP or Model Context Protocol, autonomous workflows, multi-agent systems, LangChain or Ollama, agent memory, or when the task involves giving a model the ability to act rather than only to answer.