shekohex
UserMy AI Agents Home
Categories
Indexed Skills (37)
memory
Browse and recall OpenCode local memory stored on the user's machine: sessions, conversations, and project context. Use immediately when the user asks to check history, previous sessions, past chats, what did we do before, last time, session history, recall, memory, remember, prior work, previous context, or have we done this before. Auto-trigger proactively when resuming work, continuing a project, referencing prior decisions, debugging repeated issues, or any follow-up where earlier OpenCode context may help. This means OpenCode local history specifically, not ChatGPT/Claude cloud history, generic web search, or unrelated product memory systems. Do NOT use for fresh tasks with no relevant history, or when current files/git already answer the question.
executor
Load before using the `execute` tool, external systems, and configured MCP/OpenAPI/GraphQL integrations. Use `tools.search({ ... })`, then `tools.describe.tool({ path })`, then call the full `tools.<namespace>.<tool>(args)` path.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
write-a-skill
Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.
gsd-oc-select-model
Interactive model selection workflow with paginated navigation. Use when users want to select a model interactively - guides them through provider selection then model selection using the question tool with pagination support for large lists.
upgrading-to-upstream-pi
Upgrade this agent wrapper to a newly released upstream earendil-works/pi version, including package versions, lockfile, release manifest, patch-package patches, verification, commit, and push. Use when user asks to upgrade Pi, update to a new Pi release/tag/version, regenerate Pi patches, or handle repeated upstream Pi release bumps.
creating-goals
Create or update durable agent goals by turning user intent into a polished, outcome-first closed feedback loop prompt. Use when the user asks to create, update, write, refine, or install a goal, or mentions goal prompts, closed feedback loops, goal tool usage, or goal instructions.
dynamic-workflows
Orchestrate multi-agent workflows using the `workflow` tool with fan-out/fan-in, adversarial verification, tournaments, and loops. Use when the user asks for workflows, multi-agent orchestration, fan-out, parallel subagents, dynamic workflows, or needs to decompose large tasks into coordinated subagent runs.
babysit-pr
Babysit a GitHub pull request after creation by waiting for compact PR events, handling CI/review blockers, retrying likely flaky failures, and continuing until the PR is ready, merged, closed, or blocked on user help. Use when the user asks to monitor, watch, babysit, keep an eye on, handle PR CI, or respond to PR review feedback.
brainstorming
Use when creating or developing, before writing code or implementation plans - refines rough ideas into fully-formed designs through collaborative questioning, alternative exploration, and incremental validation. Don't use during clear 'mechanical' processes
creating-issues
Create GitHub issues with `gh` in the user's concise, emoji-free style. Use this whenever the user asks to create, file, open, or report a GitHub issue or bug report, including phrases like "create a GitHub issue for that", "file an issue", "create a GitHub bug report", "report this bug on GitHub", or "open an issue". Infer the target repo automatically, strongly prefer the upstream parent repo over the current fork when working in a forked checkout unless the user explicitly wants the fork, inspect the last 5-7 authored issues to mirror tone and structure, prefer any repo-provided issue template, choose appropriate repo labels, attach the issue to the right project when relevant, and populate project fields such as status, priority, and the active sprint or iteration from recent precedent when clear.
creating-pull-requests
Create GitHub pull requests with `gh` in the user's concise, emoji-free style. Use this whenever the user asks to create, open, publish, or send a PR or pull request, including phrases like "create a new PR", "create a new pull request", "open a PR", "publish this branch", or "create a draft PR". Infer the upstream repo and base branch automatically, inspect the last 5-7 authored PRs and issues to mirror tone and structure, verify relevant local quality checks before opening, prefer any repo-provided PR template when present, and produce a brief high-signal PR with validation, review guidance, optional demo notes, optional Mermaid diagrams, and closing issue references when relevant.
dispatching-parallel-agents
Use when facing 3+ independent failures that can be investigated without shared state or dependencies - dispatches multiple Claude agents to investigate and fix independent problems concurrently
extreme-programming
Use when pair programming with humans - enforces XP values (communication, simplicity, feedback, courage, respect) to deliver high-quality software; push back on YAGNI violations regardless of seniority or sunk cost
frontend-design
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
grill-with-docs
Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.
librarian
Cache and refresh remote git repositories under ~/.cache/checkouts/<host>/<org>/<repo> so future references can reuse a local copy. Use this skill when the user points you to a remote git repository or you encountered a remote git repo through other means.
long-running-agent
Use when asked to build an entire project end-to-end, manage multi-milestone autonomous development, "build this project", "implement this end-to-end", or orchestrate long-running development that spans hours or days without human intervention
receiving-code-review
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation
requesting-code-review
Use when completing tasks, implementing major features, or before merging to verify work meets requirements - dispatches superpowers:code-reviewer subagent to review implementation against plan or requirements before proceeding
root-cause-tracing
Use when errors occur deep in execution and you need to trace back to find the original trigger - systematically traces bugs backward through call stack, adding instrumentation when needed, to identify source of invalid data or incorrect behavior
skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
subagent-driven-development
Use when executing implementation plans with independent tasks in the current session - dispatches fresh subagent for each task with code review between tasks, enabling fast iteration with quality gates
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes - four-phase framework (root cause investigation, pattern analysis, hypothesis testing, implementation) that ensures understanding before attempting solutions
test-driven-development
Use when implementing any feature or bugfix, before writing implementation code - write the test first, watch it fail, write minimal code to pass; ensures tests actually verify behavior by requiring failure first
testing-anti-patterns
Use when writing or changing tests, adding mocks, or tempted to add test-only methods to production code - prevents testing mock behavior, production pollution with test-only methods, and mocking without understanding dependencies
thermo-nuclear-code-quality-review
Run an extremely strict maintainability review for abstraction quality, giant files, and spaghetti-condition growth. Use for a thermo-nuclear code quality review, thermonuclear review, deep code quality audit, or especially harsh maintainability review.
using-gemini-cli
Run the `gemini` CLI in headless mode for one-off prompts, model-directed Gemini tasks, structured output capture, and follow-up messages against prior Gemini sessions. Use when the user asks to use Gemini CLI, mentions `gemini -p`, wants Gemini in non-interactive mode, wants a `flash` or `pro` Gemini run, needs project-scoped session discovery, or wants to continue an older Gemini session from the terminal instead of the interactive TUI.
using-superpowers
Use when starting any conversation - establishes mandatory workflows for finding and using skills, including using Skill tool before announcing usage, following brainstorming before coding, and creating TodoWrite todos for checklists
verification-before-completion
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
verify-this
Verify a claim with fresh local evidence. restate it falsifiably, capture baseline and treatment, compare artifacts, and return VERIFIED, NOT VERIFIED, or INCONCLUSIVE. Use this skill when the user asks to verify a claim or say verify this, prove it works, or did this fix it.
writing-plans
Use when design is complete and you need detailed implementation tasks for engineers with zero codebase context - creates comprehensive implementation plans with exact file paths, complete code examples, and verification steps assuming engineer has minimal domain knowledge
agent-device
Automates Apple-platform apps (iOS, tvOS, macOS) and Android devices. Use when navigating apps, taking snapshots/screenshots, tapping, typing, scrolling, extracting UI info, collecting logs/network/perf evidence, or planning agent-device CLI commands.
executing-plans
Use when partner provides a complete implementation plan to execute in controlled batches with review checkpoints - loads plan, reviews critically, executes tasks in batches, reports for review between batches
git-commiting
Read this skill before making git commits; use this skill when the user asks you to commit the code, or before pushing or making a new github pull request
webfetch
Fetch web pages and convert them to clean Markdown using markdown.new. Use when user wants to fetch a URL as markdown, convert HTML to markdown, or extract clean text from websites.
grill-me
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.