Design or review an AI agent harness: context assembly and compression, perception tools, action tools, constraints, verification, correction, and evals. Use for agent runtimes, tool loops, coding agents, RAG or document agents, long-running workflows, and context or tool failures; not ordinary one-shot prompting.
Help users operate the published Composio CLI to find the right tool, connect accounts, inspect schemas, execute tools, subscribe to trigger events with composio listen, script workflows with composio run, and call authenticated app APIs with composio proxy. Use when the user asks how to do something with composio…
★not rated 5
changed 3d agoA139 tokens
originalMIT
Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.
Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.
Write a self-contained continuation prompt for the next session. Use for "handoff", "what's next", "give me the next prompt", or unfinished work crossing a session boundary.
Design, critique, or improve frontend interfaces. Use for UI/UX, visual hierarchy, accessibility, responsive behavior, design systems, motion, copy, or browser-led polish; not backend-only work.
Audit a codebase read-only and produce prioritized implementation plans for another agent. Use for architecture, correctness, maintainability, performance, or test surveys.
Operate jk, @jkomyno's deterministic CLI, for identity, notifications, GitHub Actions pinning, and idea corpora. Use for any jk command or output contract.
Run user-facing, read-only reviews of diffs, PRs, plans, or specs for actionable correctness, architecture, test, and scope findings. Use before shipping or execution.
Write, edit, or evaluate technical prose in Alberto Schiabel's voice without inventing experience. Use for articles, plans, docs, talks, case studies, portfolio copy, or newsletters.
Design the loss function and harness for a long-running optimization run. Use for autonomous product distillation, holdout evaluation, or agent improvement loops.
Remove low-evidence code and formulaic prose while preserving intent, or install the anti-slop Oxlint policy. Use for no-slop requests and weak TypeScript contracts.
Draft a technology conference CFP package: title, abstract, outline, speaker bio, and organizer notes. Use for talk proposals and conference submissions.
Draft or revise an evidence-backed pull request body for a non-Composio repository. Use for PR descriptions or templates; defer Composio projects to their repository guidance.
Keep a pull request green and mergeable. Use for "fix CI", review feedback, CodeRabbit, failing-check links, or "babysit this PR"; stop at merge-ready or a human decision.
Research a question from high-trust primary sources and save the findings as Markdown in the repository. Use for documentation or API fact-finding.
★not rated 5 2d agoA30 tokens
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: