Showing 30751–30780 of 42323 skills
Run the two blind pre-merge reviews from delivery rule 14 on a PR — a spec review blind to the rules and a code review blind to the issue, each by a fresh subagent reading only the inputs directory prepared for it — then have a third fresh subagent verify every blocking finding and sort the nits, and hand back a fixed report block. Never edits code. Use when the deliver orchestrator reaches review, or when someone says "review PR N" or "/review N".
callstackincubator/apex-software-factory
Inspect the live Apex Factory app through Rozenite for Agents — network requests, console messages, the React component tree and renders, performance traces, heap snapshots, and the app's own tools — on a debug build with Metro started by .agents/scripts/device-lane.sh metro start. Use to collect evidence for a failing simulator check or a bug report, or when someone asks to debug the running app.
callstackincubator/apex-software-factory
Run a spec session on a GitHub issue — turn a request:new into a feature spec, write or amend its business sections with the product owner, write the Technical spec with a developer (ending with a fresh spec-checker subagent's report, fixed before the ready label), or (unattended, from deliver) split a feature:ready feature into task sub-issues from the slices its Technical spec lists. Always reconciles comments posted since the body was last edited before writing anything. Use when someone says "spec session for
callstackincubator/apex-software-factory
Triage a bug:triage issue — reproduce it as a failing test, find the root cause, and post a report proposing the simplest fix, without fixing it. Loaded by the triager subagent. Use when someone says "triage
callstackincubator/apex-software-factory
Prove a PR's simulator Done when lines and e2e: replay scripts on an iOS simulator — an Expo Simulators session running the PR's lane binary, started through .agents/scripts/device-lane.sh, checked with agent-device (open, snapshot, wait, is, replay) and shortcut with appduct — and hand back pass, fail, busy or unavailable. Always releases the session. Loaded by the mobile-verifier subagent. Use when the deliver orchestrator delegates simulator verification, or when someone says "verify #N on the simulator".
callstackincubator/apex-software-factory
After a delivery PR merges, check on an Expo Simulators iOS session that its OTA update actually landed — wait for the OTA publish run of the merge commit, start a session on the production build with the merge's native fingerprint, make the app check now with appduct's ota_check_now, apply the staged bundle by tapping the update banner with agent-device, and confirm the bundle id and the feature on screen. Always releases the session. Loaded by the ota-verifier subagent. Use when the deliver orchestrator hands over a merged PR, or when someone asks "did the update for
callstackincubator/apex-software-factory
Use when creating or changing agent instructions or their activation behavior. Skip edits that only correct spelling or formatting.
MVPavan/via
Use for Beads dependencies, intake states, recovery, workstream tracking, or unfamiliar task-lifecycle operations. Routine tracking uses AGENTS.md and runtime context.
MVPavan/via
Use when material scope or behavior decisions need exploration, or the user requests brainstorming. Handle a single routine ambiguity directly.
MVPavan/via
Use for substantive code review or to verify and act on review findings. Routine final diff inspection is not a separate review workflow.
MVPavan/via
Use for substantive module-interface, encapsulation, or testability decisions. Skip ordinary code explanations and cosmetic refactors.
MVPavan/via
Use when integrating supplied discussions or decisions into a self-contained next version of existing design documents.
MVPavan/via
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. Trigger on ADR phrases (record, write, supersede an ADR).
MVPavan/via
Use to execute tracked multi-step work, a standalone plan, or a roadmap phase. Clear bounded implementation can proceed directly under AGENTS.md.
MVPavan/via
Use when explicitly requested, or only when a consequential decision hinges on an unresolved foundational assumption that ordinary inspection or the active workflow cannot settle. Skip routine edits, straightforward fixes, settled implementation plans, and generic requests to think critically.
MVPavan/via
Use when the user explicitly requests a grilling interview to resolve or challenge material decisions. For a written critique, review the artifact directly.
MVPavan/via
Use for requested standalone HTML documents or interactive explanations when that format materially helps. Skip agent-facing files, Markdown-native deliverables, and ordinary short answers.
MVPavan/via
Find evidence-backed architecture improvements within a named area or requested codebase survey.
MVPavan/via
Use only for an explicitly requested council of named models or agents with a judge. Ordinary review and brainstorming do not need a council.
MVPavan/via
Use for measured latency, throughput, memory or cost problems, profiling, or explicit resource targets. Skip prose optimization and speculative tuning.
MVPavan/via
Use only when the user explicitly requests a perspective council or war room. Answer ordinary decisions and critique requests without a council.
MVPavan/via
Use when settled work needs a durable implementation plan, dependency graph, or workstream decomposition. Skip planning ceremonies for clear bounded tasks.
MVPavan/via
Use when a runnable throwaway experiment can resolve a specific design, state-model, algorithm, or UI question.
MVPavan/via
Use when a decision needs multiple sources gathered and weighed. Answer a single factual lookup directly from an appropriate source.
MVPavan/via
Use for material changes to trust boundaries, sensitive data, permissions, dependency execution, or untrusted inputs, and for requested security reviews.
MVPavan/via
Show the current topic visually, inline in the conversation. Use when the user asks to see the shape of code, a flow, an architecture, or a change ("/show-me", "show me", "sketch it", "draw the flow", "what does the structure look like"). For a rendered standalone document, use html-artifact instead.
MVPavan/via
Use when unsure which skill or command fits a task, when two seem to overlap, or when the user asks what skills or commands exist, what one does, or which to use.
MVPavan/via
Use when a failure cause is unclear, an attempted fix failed, or flaky behavior needs investigation. Handle obvious localized errors with a focused reproduction and fix.
MVPavan/via
Teach a concept, walk through a session, or conduct an explicitly requested ongoing course.
MVPavan/via
Use for risky behavior changes, regression proof, legacy characterization, or substantive test design. Trivial test renames do not require a test-first workflow.
MVPavan/via