Set up AI to work for you
71 free AI skills and workflows for this job.
- Ad prompt comparison review: Paste old and new prompt ads. Get a blind test, a decision and a score out of 100.
- Agent action approval reviewer: Share a proposed agent action. Get its scope, risks and missing evidence checked before execution.
- Agent answer feedback designer: Share an agent task. Get checks and feedback loops to improve the quality of its answers.
- Agent context setter: Describe your project. Get the right rules and files loaded so your AI stays accurate through long sessions.
- Agent goal writer: Give it a vague job. Get a goal an AI agent can finish, with proof of done, limits and stop rules.
- Agent instructions auditor: Point it at your repo. Get a short, checked AGENTS.md with real commands and no guesses.
- Agent loop designer: Say the job you want repeated. Get a bounded AI loop with checks, stop rules and handoffs.
- Agent OWASP risk checker: Share your agent code. Get findings against the OWASP agent risks, with evidence and fixes.
- Agent plugin integrity checker: Share your agent tools and packages. Get file checks and a record of supply chain risks.
- Agent prompt writer: Paste a rough request. Answer a few questions. Get one clear prompt to give an AI agent.
- Agent readiness dashboard builder: Point it at your repo. Get a dashboard showing what stops coding agents from working well.
- AgentRC instruction file writer: Point it at your repo. Get project rules for coding agents, with a preview before files change.
- AI agent architecture planner: Describe the agent you need. Get a system design and the trade-offs to settle before building.
- AI agent control designer: Share how your agent uses tools. Get controls for permissions, trust and risky actions.
- AI agent workflow designer: Describe the job. Get a multi-agent design with handoffs, failure handling and cost limits.
- AI app evaluation planner: Share your AI app and goal. Get test cases and a way to compare changes with real results.
- AI build-or-buy advisor: Describe an AI plan. Get an API vs fine-tune call, AI risk class under EU rules and cost break-even.
- AI coding agent rules writer: Describe your project. Get a plan file and agent rules for working one test at a time.
- AI memory habit setter: Connect any memory tool. Get clear rules for when your AI recalls and saves decisions, so it stops repeating itself.
- AI model training debugger: Describe your model, data and what goes wrong. Get a check order that finds silent bugs before tuning.
- AI prompt editor: Share your rough prompt. Get a clearer version with the task, facts and requested result stated plainly.
- AI task brief writer: Paste a vague job. Get a goal an AI agent can finish, with proof of done and stop rules.
- AI task handover writer: Describe work to pass on. Get a clear handoff to another AI agent, with checks along the way.
- AI task opportunity review: Describe a job and who does it. Get the tasks that drain or endanger people and where AI can help.
- AI workflow instructions planner: List your AI workflow steps. Get clear instructions for passing facts and results between them.
- AI workflow quality investigator: Share a workflow that disappoints. Get its weak points ranked, with causes and practical fixes.
- AI writing workflow planner: Describe how your team writes. Get what AI should draft, what people own, and rules that keep your voice.
- Backup and recovery planner: Describe your systems. Get how much loss you can take, a backup plan and a restore drill to run.
- Check an AI app improvement: Share your AI app and proposed change. Get evaluation criteria, feedback loops and an evidence-based comparison plan.
- Claude memory curator: Point it at Claude Code memory. Get what it learned reviewed, kept rules moved up and stale notes cut.
- Claude power-user coach: Tell it your work. Get lessons and exercises that make you faster with Claude, one step at a time.
- Coding agent readiness criteria planner: Share your repo's needs. Get scoring rules that fit the project and show which checks matter.
- Content to workflow builder: Paste a long article, transcript or book notes. Get the method as a workflow, with a skill for every step.
- Focus-friendly answer style setter: Turn it on once. Get every reply with the next action first, short numbered steps and time estimates.
- Folder guide writer: Pick a folder. Get a short CLAUDE.md that tells future AI helpers how that folder works.
- Full answer enforcer: Ask for a full file or list. Get every line, with no placeholders, skipped parts or half answers.
- Give AI tasks and keep what the job quietly does: Give it a job. Get the tasks AI takes on, what each hides, safe handoffs and a failure check.
- Goal-to-task loop runner: Give it a goal and a folder of skills. Get a checked task plan that runs one task at a time.
- Local AI model setup: Say what hardware you have. Get Ollama and Open WebUI running local models, tuned for your graphics card.
- Long chat failure diagnoser: Describe how your AI gets worse in long chats. Get the likely cause and a fix to try.
- Long chat summariser: Give it a long AI session. Get a structured summary that keeps files, decisions and next steps.
- Make your repo agent ready: Share your repo. Get a readiness dashboard, agreed scoring rules and project instructions for coding agents.
- Marketing prompt tester: Give it your prompts and test cases. Get scored comparisons of each version and a saved version history.
- MATLAB installer: Name the MathWorks products. Get them installed from the command line, the same way each time.
- MATLAB product lister: Point it at a MATLAB install. Get a list of every product and add-on package in it.
- MCP handler security reviewer: Share your MCP code. Get risks traced through tool handlers, with evidence and fixes.
- MCP protocol release tester: Share your MCP server. Get a real session check of its tools, errors and advertised features.
- MCP settings risk checker: Share your MCP settings. Get risks in permissions, trust and server access, with fixes.
- Metric-driven file improver: Pick a file and a score. Get an agent that edits, measures and keeps only changes that raise the score.
- Multi-agent workflow planner: Give it a big task. Get a plan that splits it among AI agents with clear owners, handoffs and checks.
- Parallel work coordinator: Give it a large task. Get it split across helper agents with clear ownership, then merged into one checked result.
- Project memory builder: Give it a bloated project notes file. Get a lean version, plus a way for your AI to save lasting facts.
- Project memory organizer: Point it at your project. Get small context files per folder so your AI reads less and costs less.
- Prompt and skill tuner: Give it a prompt or skill and test cases. It tests, fixes and re-tests until the checks pass, within four rounds.
- Prompt change manager: Keep prompts in production. Get versioning, tests of changes, a registry and safe rollbacks.
- Prompt fixer: Give it a prompt that gives weak or uneven answers. Get a tighter version that works more reliably.
- Prompt improver: Give it a rough prompt. Get it rebuilt into a clear, structured prompt using a fitting framework.
- Prompt safety and bias reviewer: Paste your prompt. Get risks in its wording, likely failure cases and a safer draft.
- Prompt tester and tuner: Give it a prompt and test cases. Get scores for each version and a better prompt backed by results.
- Prove your new ad AI setup beats the old one: Give it a story ad test plan. Get handoff cards that keep facts whole, then a blind old-versus-new score.
- Recurring marketing routine builder: Pick an outcome. Get a loop that runs weekly or daily, checks itself and stops when it should.
- Repeatable AI loop designer: Describe a recurring job. Get a loop with triggers, checks, stop rules and handoffs.
- Request acceptance brief: Turn any concrete request into a scope, input list, named deliverables and acceptance checks.
- Request acceptance checker: Check generated files against every requirement, repair bounded failures and package an honest handoff.
- Request deliverable drafter: Use an agreed request brief to create its named files, with source evidence and explicit gaps.
- Routine task automation planner: List what your team does each week. Get the chores worth handing to a tool, with steps a new hire can run.
- Short reply mode: Switch it on. Get much shorter AI replies that drop filler but keep the facts, to save tokens.
- Task automation builder: Name a repetitive task. Get a script, schedule or trigger that does it for you.
- Token saver for coding agents: Turn it on for coding sessions. Get fewer wasted tokens from big reads, noisy logs and repeat reads.
- Tune a prompt with real tests, then keep versions: Give it a rough prompt. Get a clear version, scores on your tests and safe rules for changing it.
- Workplace shorthand memory builder: Share your nicknames and codes. Get a memory so Claude knows who and what you mean in short requests.