Security first
Safety Radar beta now; performance benchmarks, private registries, and runtime skill recommendations next.
Agent Skill Kits
Skip the random browsing. SkillRadar Kits group recommended agent skills by workflow, show the safety profile up front, and let you watch the whole bundle for risky changes.
4 skills · avg trust 68
Make a coding agent better at planning, API review, tests, and shareable diffs.
A practical starting stack for software builders who want their agent to reason about architecture, review APIs, create test plans, and return cleaner code changes.
4 skills · avg trust 84
Give your agent web research, community search, repo issue search, and durable notes.
A research-oriented bundle for agents that need to collect evidence, search communities, inspect repo history, and preserve findings in a useful knowledge base.
4 skills · avg trust 90
Help an agent improve landing pages, ecommerce funnels, sourcing, and experiments.
A revenue-focused kit for agents working on ecommerce, landing pages, sourcing decisions, recovery flows, measurement, and growth experiments.
4 skills · avg trust 49
Review prompts, security-sensitive changes, advisories, and API risk before automation acts.
A defensive kit for teams that want their agents to spot prompt-safety problems, triage security issues, and review risky interfaces before changes ship.
4 skills · avg trust 67
Extend an agent with autonomy, MCP building, model routing, and command/plugin workflows.
A higher-leverage kit for people turning agents into operating systems: more autonomy, integrations, routing, and extensibility with clear risk visibility.
Why this matters
Install recommendations are useful once. Watched recommendations are useful every week. SkillRadar helps users pick a starting stack, then keeps monitoring the skills for command, credential, domain, installer, and risk-signal changes.