tighten
01Skill Claude CodeCodex
../../../.claude/skills/tighten/SKILL.md.
Skill Claude CodeCodex
../../../.claude/skills/tighten/SKILL.md.
Plugin Claude Code
AndThen – structured workflows for agentic development: clarify → spec → plan → execute → review.
Hook Claude Code
Runs before the agent uses a tool for Bash tool calls, executing block-dangerous-commands.py via python3. From IT-HUSET/andthen.
Settings file Claude Code
Agent settings declaring 1 hook event (PreToolUse) and 41 allowed tools.
Instructions file CodexOpenCode
AGENTS.md instructions for IT-HUSET/andthen, covering ai coding agent instructions for working with andthen, project overview, core concepts, repo map and project document index.
Instructions file
Claude Code instructions for IT-HUSET/andthen, a project described as: Coding agent toolkit with workflows, guidelines, and plugins for Claude Code and other agents.
Plugin Claude Code
AndThen workflow system: structured agentic development from requirements to shipped code.
Agent
Fetches up-to-date, version-specific documentation and code examples for libraries, frameworks, and services. Use when implementation needs an API reference, configuration option, migration guide, deprecation, or official example. Returns distilled conclusions, not page dumps.
Agent
An expert web researcher using advanced search techniques and synthesis. Masters search operators, result filtering, and multi-source verification. Handles competitive analysis and fact-checking. Use PROACTIVELY for deep research, information gathering, or trend analysis.
Agent
Agent-workflow reviewer for AndThen review councils – catches skill-vs-agent-type confusion, gate-skipping loops, and prompt portability breaks across Claude/Codex/generic installs. Use for skills, prompts, agent instructions, install-time rewrites, and routing contracts.
Agent
Architecture reviewer for AndThen review councils. Use for coupling, boundaries, abstractions, maintainability, resilience, and project architecture alignment.
Agent
Correctness reviewer for AndThen review councils. Use for code behavior, edge cases, data flow, error handling, and tests.
Agent
Critic reviewer for AndThen reviews. Use when a review needs an adversarial finding pass (red-team / skeptic review) that attacks assumptions, unhappy paths, hidden coupling, guessed behavior, and incomplete wiring. Do not use for the findings-filter/devil's-advocate pass; use the Devil's Advocate findings-filter…
Agent
Findings-filter reviewer for AndThen review councils. Use after specialists have produced findings to validate, downgrade, withdraw, or dispute weak findings without adding new ones.
Agent
Product and requirements reviewer for AndThen review councils. Use for user value, acceptance criteria, scope fit, requirement gaps, and feature intent.
Agent
Project standards reviewer for AndThen review councils. Use for local conventions, repo guidelines, naming, maintainability, documentation drift, and agent-instruction compliance.
Agent
Security reviewer for AndThen review councils. Use for auth, authorization, trust boundaries, secrets, injection, supply chain, LLM or agent flows, and other exploitability-focused review.
Agent
Final synthesis challenger for AndThen review councils. Use after Devil's Advocate filtering to merge, reframe, downgrade, or reject surviving findings before the final report.
Agent
Test strategy reviewer for AndThen review councils. Use for test intent, coverage gaps, weak assertions, missing edge cases, and verification quality.
Skill Claude CodeCodex
Architecture design and analysis – review, decomposition, trade-off analysis, ADRs, CUPID/DDD guidance, fitness functions, strategic design, event storming. Seven modes, runnable singly or chained (--mode review,fitness). Trigger on 'architecture review', 'bounded context', 'trade-off', 'write an ADR'.
Skill Claude CodeCodex
Discovery & Ideation for requirements at feature or product scope – systematic discovery of gaps, edge cases, scope boundaries, and alternatives the user hadn't considered. Trigger on 'clarify requirements', 'what are the requirements', 'product vision'.
Skill Claude CodeCodex
End-to-end browser testing for web apps – discover journeys, run interactive tests, validate responsive behavior. Trigger on 'test the app end to end', 'e2e test', 'browser test'.
Skill Claude CodeCodex
Create high-visual-quality Excalidraw diagrams where structure and design do the explaining. Trigger on 'draw this architecture', 'workflow diagram', 'create an Excalidraw diagram'.
Skill Claude CodeCodex
Execute a fully-specced implementation plan bundle (every story already has a FIS) – fixed per-story pipeline plus a final gap review. Supports Agent Teams (--team) and per-story worktree isolation (--worktree). Trigger on 'execute this plan', 'run the plan', 'run as team'.