withkynam/vibecode-pro-max-kit

Agent Browser automates web interactions for AI agents using compact snapshots and refs. Use it for long autonomous sessions, self-verifying workflows, video recording, and cloud browser testing.

Agent Strategy Compare evaluates sequential, parallel, workflow, and agent-team execution for a fan-out task. It scores seven signals, shows agent count math and cost guards, and recommends a strategy.

Audit Context verifies context routing, shared-skill discoverability, and Claude/Codex wiring. Use it when context docs, skill surfaces, or routing rules move, split, or drift.

Audit Plans reviews active plan files for staleness, completion, and routing truth. Use it when cleaning up plans, reconciling active work, or archiving completed artifacts.

Audit VC verifies agent harness wiring across Claude, Codex, README.md, and protocol files. Use it when agents, skills, or development-protocol files move, split, or drift.

Closeout Packet Generator creates the post-EXECUTE closeout packet for a completed plan or phase. It summarizes readiness, drift, verification, cleanup, and the next valid state.

Context Discovery loads the relevant process and context files for a task at the start of an agent session. It routes by domain and lists feature files with full paths before work begins.

Debugging & System Investigation guides systematic root-cause analysis for bugs, test failures, CI issues, performance problems, and other unexpected behavior before applying fixes. It also covers log analysis, verification, and frontend checks when needed.

Documentation Discovery finds library and framework docs through Context7 llms.txt lookup, with scripts for fallback repository and topic searches. Use it when you need API docs, setup details, or recent library information.

Feasibility Test runs a one-shot probe to check whether an unverified runtime, library, or external mechanism actually works as assumed. Use it before locking a design when the answer depends on live behavior.

Generate Context updates the project's authoritative context files at process/context/all-context.md and related group files. Use it when repository context is missing, stale, or contradicted by code.

Generate Plan creates or updates implementation plans in the repo’s SIMPLE or COMPLEX format. Use it when turning an idea, PRD, or approved direction into a saved plan artifact.

Intent Clarifier scores request ambiguity and routes it into the right phase. It generates structured, option-rich questions when a request needs clarification before delegation.

Phase Program Generator creates kickoff artifacts for a multi-phase program, including an umbrella plan, Program Goal Charter, phase stubs, and the per-phase inner-loop reference. Use it when work spans several dependent phases that need validation gates and parallel planning.

Plan Discovery finds plan artifacts related to the current task, including same-feature plans, other active feature plans, and general plans. Use it before research, validation, execution, or updates to see what was tried, deferred, or referenced.

Problem Solving Techniques helps you choose and apply systematic methods when coding gets stuck. Use it for complexity spirals, repeated patterns, assumption traps, or uncertainty about scale and production readiness.

Review Situation summarizes read-only branch, worktree, and active plan state for handoff and resume checks. Use it when you need a current repo orientation without changing anything.

Risk Evidence Pack defines the manual-first artifact set required before finalizing high-risk work. Use it to capture risk classification, code evidence, verification results, review decisions, and adversarial checks.

Scenario Explorer generates edge cases and test scenarios by decomposing a feature across 12 dimensions. Use it before implementation or testing to surface risks, failures, and coverage gaps early.

Spec Generator creates or updates a plain-language product-discovery SPEC for user review. Use it after research and intent are known, before planning or implementation begins.

Vc Auto Research runs a find-gaps, fix, and recheck loop for specs, tests, docs, and UX hardening. Use it when you want repeated verification until no gaps remain or a metric target is reached.

Vc Autopilot emits, validates, and resumes the provisional goal block used in Autopilot Mode. Use it when an orchestrator needs to detect a pasted goal block or write a 9-field artifact.

Vc Publish pushes harness improvements from the current development repo into the remote kit repo. It compares managed files, shows the diff, and updates the kit version before pushing.

VibeCode Frontend Design creates polished frontend interfaces from screenshots, videos, or descriptions. Use it to replicate UIs, build 3D or immersive web experiences, and avoid generic AI-generated layouts.

VibeCode Predict runs five expert personas to review a proposed change before implementation. Use it to catch architectural, security, performance, and UX issues before major or risky work.