quiver

5-phase research pipeline that dynamically detects the project's tech stack from manifests and lockfiles, validates against current documentation via context7 MCP, and flags deprecations before they become bugs. Every recommendation is version-checked and source-cited.

Cross-references the diff against project memory, git history, and past decisions to surface institutional knowledge, churn patterns, and recurring issues that contextualize the review.

Evaluates structural integrity via context7-driven convention discovery, diff manifest-aware boundary analysis, and pattern compliance grounded in the project's actual codebase -- not abstract ideals.

Transport adapter that runs the OpenAI Codex CLI to perform code review on the orchestrator-provided diff and emits Codex's structured findings verbatim. Does not review or interpret; relays only.

Evaluates code changes for developer experience quality across discoverability, error messages, debugging experience, and automation-readiness for both human developers and AI agents.

Systematic path tracer that finds logic defects by tracing each changed function's inputs through branches to outputs -- verifying correctness at every step rather than scanning for known bug patterns.

Adversarial 7-phase security auditor with multi-platform mobile coverage (Flutter, Kotlin/Android, Swift/iOS), prompt-vs-code awareness for LLM plugin contexts, and supply chain analysis across CI scripts and dependency resolution.

Constructs concrete failure scenarios by stressing assumptions, fracturing component interactions, and building cascade chains -- each finding is a narrative showing trigger, propagation, and failure state.

Evaluates whether tests actually prove code works by analyzing assertion strength, regression detection power, and risk-based coverage gaps -- asking 'would this test catch a real bug?' not just 'does a test exist?'

Detects wasted effort in diffs: unnecessary files, dead code paths, redundancy with existing codebase utilities, over-engineered abstractions, and ceremony the framework already handles.

Browser-based visual brainstorming companion for showing mockups, diagrams, and visual options. Use when brainstorm topics involve UI/UX, layout, architecture diagrams, or any content better understood visually.

Execute a work plan or specification systematically -- read the plan, set up a branch, implement tasks with continuous testing, commit incrementally, and ship a PR. Use when you have a plan file, spec, or task list ready to execute.

Explore ideas, compare approaches, and produce a validated spec before planning.

LSP-first code navigation with grep fallback. Provides agents with a reusable strategy for semantic code exploration -- goToDefinition, findReferences, documentSymbol -- with automatic fallback to grep when LSP is unavailable or returns empty results.

Generate a Conventional Commits message, commit, and optionally push to remote.

Scaffold a new Claude Code agent with smart defaults from a natural language description or interactive Q&A.

Generate or rewrite an AGENTS.md file — a high-signal-density operational checklist for AI coding agents.

Create a GitHub pull request from the current branch. Use when the user says 'create a pr', 'open pr', 'push and open a pr', 'create pull request', or wants to open a PR from their branch.

Delete all handover files for the current project to completely reset session history.

Delete the most recent handover file to remove the last session's context.

Summarize the current work state and prepare a handover note for the next session.

Load the most recent handover note from the previous session into context.

This skill should be used when delegating a task to multiple agents, assembling agent teams, or coordinating parallel agent execution. It applies when the user asks to run agents on a task, orchestrate work across specialists, or when a complex request benefits from decomposition into agent-handled subtasks.

Create a structured implementation plan with parallel agent research before coding.

Diagnose and fix a broken skill by analyzing its structure, verifying API references against current docs, and applying targeted repairs.

Run a multi-agent code review (code quality + security audit + architecture analysis) with synthesized findings. Pass --with-codex for cross-model coverage via the OpenAI Codex CLI.

1 hook across 1 event