github-solution-research
Find open-source solutions on GitHub. Use when facing a bug, integration failure, dependency issue, unclear API usage, implementation blocker, or tool capability need.
github-triage
Triage GitHub issues. Use when the user wants to triage incoming issues, prepare issues for an autonomous agent, or move an issue between workflow states.
handoff
Snapshot the current session into a resumable handoff artifact so a cold session, agent, or person continues without replaying context. Use when the user says "write a handoff", "hand off", "snapshot this session", "I'm running low on context", "pause and pack this up for the next session", or "resume this later".
ideate
Generate grounded, divergent ideas from the codebase. Use when the user says "let's brainstorm", "what should we build", or "any ideas for".
improve-architecture
Surface deepening refactors that turn shallow modules into deep ones. Use when the user asks to improve architecture, find refactor candidates, raise testability, or make a codebase more agent-navigable.
incremental
Use when a change spans multiple files, or you're about to write a lot of code at once.
inits
Creates or improves an AGENTS.md file. Use when onboarding to a repo, or capturing conventions, constraints, and rationale that are expensive to rediscover.
llm-self-loop
Restructures human-gated workflows into autonomous LLM loops with file-based outputs. Use when a task needs a button click, dashboard check, or human verdict inside its iteration loop.
loop-me
Use when the user wants to design a recurring personal or team workflow: capture loops, write specs in workflows/*.md, and run a stateful askme session until the spec is implementer-ready. Triggers: "loop me", "design a workflow", "automate this recurring task", morning routine, inbox processing.
memory-clean
Use when the user asks to audit memory or find stale or duplicate memories.
memory-sanitize
Use when sanitizing memory for sharing, redacting PII, or scanning for credentials.
memory-update
Use when the user says save this to memory, or scan this session for memories.
minimalism-driven
Build the smallest version that fully does the job and makes its intent obvious — the right amount, not the fewest lines. Use when scope creep, speculative abstraction, or unneeded surface is a risk; before adding a helper, wrapper, config key, or dependency; or when the user says "minimal", "DRY this up", or "no gold-plating".
mutual-sync
Use when user, agent, and codebase may hold different pictures of the current state: before substantive work after a gap, or mid-session when parties talk past each other. Triggers: "mutual sync", "sync up", "are we on the same page", stale user claims about the code, refuted agent assumptions.
observability
Instrument code with structured logs, metrics, and traces. Use when adding telemetry, setting up or reviewing alerting rules, shipping a production feature, or diagnosing opaque production issues.
onboard
Use when the user asks to be onboarded, oriented, given a tour, or shown where to start in a repo.
optimize
Use when the user asks to optimize something, reduce allocations, or fix a performance regression.
parallel-launch
Use when the request has multiple independent sub-tasks, cross-domain research, or parallelizable file or module work.
pov
Verdict on whether to adopt, switch to, or revisit a technology, library, pattern, platform, or architecture; also a mid-session second opinion. Use for a verdict against the current project, not for neutral explainers or option-generation.
pr-merge-base
Use when PRs may conflict with each other or the base, or require ordered application.
pr-merge-temporal
Use when validating a set of PRs together on a staging branch before advancing the base branch.
pr-review
Use when the user asks to review a pull request, or wants structured quality, security, and testing feedback on a PR diff.
proof-driven
Use when using property-based testing, theorem proving, or formal proof tactics with a zero-unproven-property policy.
refactor-break-compat
Use when modernizing APIs, removing compat shims, killing feature flags, or rewriting a subsystem cleanly.
research
Use when researching a named library, framework, SDK, API, or service, or finding a migration guide.
resolve-pr-feedback
Use on a GitHub PR when addressing review comments or resolving review threads.
resolve
Use when addressing review feedback or analyzing whether a comment is valid.
source-driven
Use when writing code against a specific framework or library version, building boilerplate, implementing a framework's recommended pattern (auth, routing, forms, data fetching, state), verifying an API signature before writing framework-specific code from memory, or the user asks for a documented, verified, or "correct" implementation.
spec-driven
Use when starting a new project or feature, requirements are unclear, or the change touches multiple modules and needs a shared spec.
strategy
Use when defining product strategy, setting a north star, starting or redirecting a product, or an existing STRATEGY.md has gone stale.
strict-validation-setup
Bootstrap strict-mode tooling and per-task GOALS.md scaffolding so an agentic loop can self-verify. Use when setting up a new project, making a task self-verifying, or scaffolding goals for an agentic loop.
subagent-driven
Execute a multi-task plan. Use when the user says "execute with subagents", or hands you an ordered multi-task plan to run.
sync-docs
Detect docs-vs-code drift from a diff. Use when the user says "sync docs" or "update changelog".
taste
Apply taste judgment to an artifact or decision. Use when asking "is this slop?", "overkill?", "audit", or "taste-test this", or setting a taste register.
teach
Use when the user wants to learn a skill or concept across multiple sessions in a persistent teaching workspace at the cwd: they ask you to set up a course, design lessons, build a learning workspace, or explicitly request ongoing teaching. Not for one-off explanations or quick answers.
test-driven
Run test-driven development. Use when implementing features or fixes with TDD, or refactoring with coverage.
tests-adversarial
Write adversarial tests. Use when hardening error handling, validating boundary behavior, or hunting silent failures.
tests-purge-unneeded
Delete tests that do not catch real bugs. Use when reviewing legacy suites, investigating slow CI, sweeping tests after a refactor, or checking whether the type system already covers an assertion.
tidy
Dispatch compress operations. Use when the user says "tidy" or names a file, diff, memory, workspace, git stack, or doc to tidy.
to-questionnaire
Use when the user wants to turn a knowledge gap into an async questionnaire for someone else: identify the recipient and the needed answers, then write a discovery questionnaire to to-questionnaire-<slug>.md. Triggers: "turn this into a questionnaire", "questionnaire for X", "I need to ask someone else", async information gathering.
type-driven
Develop with refined types. Use when modeling a domain, encoding a state machine in types, hardening an API boundary against malformed input, or the user says "make invalid states unrepresentable" or "parse don't validate".
validation-first-driven
Define state machines, invariants, and temporal properties. Use when building protocols, workflows, concurrent systems, or lifecycle-heavy state.
verification-before-completion
Requires fresh, actually-run evidence before any claim that a task is complete. Use when about to tell the user a task, feature, or fix is done, complete, finished, working, or ready, or about to report a DONE or passing status.
work
Execute a plan or concrete work prompt end-to-end. Use when implementing from a plan or spec path, or a clear build request; use /debug for open-ended bugs.
workflows-driven
Drive decomposable work as a deterministic multi-subagent workflow: phased fan-out under per-task contracts with adversarial verification. Use for audits, migrations, broad research or review sweeps, work needing independent cross-checks before committing, or scale one context cannot hold.
worktree
Set up isolated git worktrees: create a new branch for fresh work, or attach a worktree to an existing branch/PR/commit.
writing-beats
Use when the user wants to build an article as a choose-your-own-adventure journey of beats, sequence raw material one move at a time, or says "write this beat by beat", "journey-style article", or "assemble beats".
writing-fragments
Use when the user wants to generate raw writing material before structuring it, has scattered ideas for an article, or says "collect fragments", "mine this topic", or "gather notes first".
writing-shape
Use when the user wants to turn a pile of fragments into a polished article, shape raw material paragraph by paragraph, or says "write this up", "turn this into a post", or "structure these fragments".
writing-skills
Reference for writing and editing skills so they run predictably — invocation choice, information hierarchy, leading words, pruning, and the failure modes that make a skill misfire. Use when authoring or refactoring a SKILL.md, or the user says "write a skill", "improve this skill", or "why does my skill fire unreliably".
Page 2 of 3 · 101 results