Turns vague software ideas into clear, implementation-ready specifications for AI coding agents, asking only the questions that matter and clearly separating your requirements from AI assumptions and engineering recommendations.
Scan AI Skills and MCP servers before installation. Detect risky scripts, permissions, dependencies, network access, credential exposure, file changes, and suspicious configurations — then generate a clear evidence-based security report.
Audit AI agents, Skills, MCP configurations, tool permissions, and coding-agent setups for prompt injection, excessive agency, credential exposure, data leakage, unsafe code execution, supply-chain risks, memory risks, missing approval controls, and runaway autonomy. Get a calibrated Security Score, Audit Confidence, evidence-backed findings, remediation steps, verification checks, and residual-risk analysis.
This Skill is built on practical expertise in AI agent safety, prompt-injection defense, source-to-sink privacy review, tool and MCP metadata integrity, memory provenance checks, autonomous action scoping, and evidence-based verification. It helps agents separate instructions from evidence before tool calls, uploads, publishing, deletion, code execution, or network output.
Turn every stack of business cards into your next opportunity. Get a structured contact list from card photos or PDFs: Chinese, English, and bilingual cards, front and back sides and duplicates merged, uncertain details flagged, Excel and CSV ready. Switch to Analyze mode to classify companies, research public information, compare business fit with your company, and learn who to contact, why, and what to say next, with a follow-up email draft. Analyzes companies, not individuals.
Create a reversible, project-local Codex copy of Claude Code agents, skills, commands, compatible hooks, MCP, and persistent GPT routing—while keeping Claude untouched.
Drive real Android phones from your desktop over wireless ADB. Capture screens, record demos with a virtual cursor that shows exactly where each tap landed, annotate screenshots with numbered callouts, smoke-test builds on real hardware, read notifications, and automate app-only flows that have no public API. Replaces the manual capture-crop-annotate-reshoot loop with scripts.
Turn recurring social, banner, and ad requests into a repeatable marketing asset production system. This download skill builds a brief intake gate, template decision, content-variable batch plan, brand QA checklist, handoff/version-control rules, and a performance feedback loop, then writes a campaign production package to journal/marketing-assets/. Built for teams whose designer is the bottleneck and whose assets drift across channels.
Audit AI agent workspaces for authority, secrets, data flow, memory, supply-chain, and evidence risks—locally and without executing target code.