Project Phase Manager is a skill published by agensi that converts complex, high-level objectives into structured, enterprise-grade execution plans. It organizes work into formal phases, milestones, and parallel workstreams, each governed by strict entry and exit criteria. The skill acts as an automated Project Management Office for both AI agents and human teams, replacing generic task lists with accountable, dependency-driven planning. It enforces professional governance standards, including critical path analysis through finish-to-start, start-to-start, and finish-to-finish relationships. It also applies risk management and evidence-based status tracking across the project lifecycle. Once installed, an agent can use this skill to decompose objectives into phased plans, map dependencies between parallel streams, define entry and exit criteria for milestones, and produce precise handoff briefs for specialized agents. It integrates agent outputs at formal verification gates, ensuring that work meets criteria before progressing. The skill is intended for orchestrating multiple agents rather than producing simple task lists. It is the wrong choice when you need a lightweight task tracker without governance, critical path logic, or formal phase gates, or when you require direct integration with external repositories, package managers, or command-line tooling, none of which are provided.

Use cases

Use it to decompose a high-level objective into formal phases with milestones and entry/exit criteria.

Use it to map finish-to-start, start-to-start, and finish-to-finish dependencies across parallel workstreams.

Use it to generate a critical path analysis for an agent-driven project.

Use it to produce handoff briefs when passing work between specialized agents.

Use it to integrate agent outputs at formal verification gates.

Use it to track project status with evidence-based risk management.

When to use it

You need structured project governance for multi-agent workflows.

You require dependency logic and critical path analysis rather than simple task lists.

You want automated handoff briefs and verification gates between agents.

Your team needs entry and exit criteria for each project phase.

You are coordinating parallel workstreams for AI agents and human teams.

When not to use it

You need direct integration with external package registries or repository commands, which are not provided.

You require environment variable configuration or runtime secrets management, which this skill does not expose.

You are looking for a simple todo list without formal governance, phases, or verification gates.

You need built-in tools for file I/O or code execution, which are not included.

Your use case demands specific third-party tool integrations not listed in the skill.