alieismy/codex-three-layer-delivery
Instructions file CodexOpenCode
Instructions for alieismy/codex-three-layer-delivery, covering agents.md — repository maintainer instructions, purpose and authority, before editing, change and mirror discipline and dynamic task state.
alieismy/codex-three-layer-delivery
Settings file Claude Code
Agent settings configuring attribution, enableAllProjectMcpServers, permissions.
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Coordinate a multi-stage, multi-document decision or design engagement across requirements, research, feasibility, solution, design, specification, writing, and review. Use when the user explicitly requests an end-to-end workflow, document package, phased delivery plan, cross-session handoff, or artifact and decision…
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Produce an implementation-ready detailed design from approved requirements and architecture decisions. Use when interface contracts, schemas, critical flows, state machines, configuration contracts, trust boundaries, errors, security, observability, concurrency, or recovery behavior must be specified for software or…
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Evaluate whether a project, product, open-source component, AI tool, infrastructure change, or candidate option is viable, and produce a decision-ready feasibility conclusion. Use when technical, economic, schedule, operational, security, compliance, and risk evidence must be compared before commitment. Route…
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Turn raw needs, meeting notes, stakeholder feedback, or business scenarios into a PRD, SRS, structured requirements, or traceability material. Use when scope, actors, rules, constraints, priorities, and acceptance must be made explicit and verifiable, including product, system, infrastructure, or AI-tool requirements.…
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Research and verify decision-critical claims using project evidence and inspected primary sources. Use for standards or policy evidence, open-source and AI-tool evaluation, API or product facts, network or system configuration research, literature review, and technical or public-affairs fact-checking. Produce…
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Independently review requirements, feasibility studies, research or tool evaluations, technical proposals, detailed designs, standards, technical articles, or public-affairs articles. Use when findings must identify location, evidence, impact, severity, remediation, and a defensible approval or publication verdict.…
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Convert approved requirements or feasibility inputs into a technical proposal, high-level design, solution architecture, outline design, or construction plan. Use when real alternatives, architecture decisions, operating boundaries, and a design-ready recommendation are needed for software, infrastructure, networking…
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Draft or revise national, industry, enterprise, internal, or project standards, specifications, guidelines, and normative clauses. Use when authority, scope, terminology, references, conformity criteria, or normative wording must be established or changed. Route an independent verdict to rd-review, and route…
alieismy/codex-three-layer-delivery
Skill Claude CodeCodex
Write source-backed technical articles, white papers, research or fact-check reports, decision briefs, and other professional explanatory documents. Use when verified evidence must be shaped for a real audience and publication or decision purpose without losing uncertainty, counterarguments, or traceability. Route…
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Project-wide conduct for Cursor Agent: language, truthfulness, independent judgment, document task modes, and output checks.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Document delivery discipline for Cursor Agent: read source material first, minimize changes, preserve intent, and convert document tasks into verifiable goals.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Cursor-native spec injection: how project rules, module rules, skills, and durable decisions should be organized.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Human-AI collaboration modes for document-delivery work, based on risk, ambiguity, authority, and verification strength.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Context management, session refresh, document maps, and search hygiene for long Cursor Agent sessions.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Research and evidence discipline: requirements clarity, official-source preference, standards rigor, evidence tiers, and optional MCP routing.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Quality gates for specialist document work, explicit delivery orchestration, and final verification.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Security and system-design baseline for Cursor Agent work: secrets, trust boundaries, compliance, auditability, and high-risk design areas.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Gotchas log and verification discipline: record recurring failure patterns and report completed, verified, unverified, and risky items.
alieismy/codex-three-layer-delivery
Cursor rule Cursor
Task routing for eight specialist rd- skills, one explicit delivery orchestrator, and their authoring-to-review relationships.