Manus adds Project Skills to its AI agent platform

Manus, the autonomous AI agent platform now part of Meta, has introduced a feature called Project Skills that lets teams curate and lock sets of reusable AI workflows at the project level.

The update, announced February 12, builds on the platform's existing Skills system by adding organizational controls for team collaboration. The feature details in this article are drawn from Manus's product announcement.^[1]

Context and background

Manus is an autonomous AI agent that can independently execute multi-step tasks such as research, code generation, and data analysis. Manus was originally developed by Butterfly Effect, a startup founded in China. The company moved its headquarters to Singapore in mid-2025 before Meta announced the acquisition in December 2025. Financial terms were not disclosed, though reports valued the deal in the $2–3 billion range.^[2][3][4]

Manus introduced its broader Skills system on January 27, 2026, allowing individual users to package successful AI workflows into reusable modules.^[5] Project Skills, announced February 12, extends that system with team-level controls. Project administrators can curate which skills are available within a workspace and lock configurations to prevent changes.

Key details

Project Skills allows users to build a curated library of AI workflows for each project, drawing from shared team skills or personal collections. When a task runs inside a project, only skills explicitly added to that project's library are available. Skills from outside the project are not triggered.

Teams can also lock a project's skill set, which Manus describes as creating a "gold-standard" workflow. This prevents other project members from modifying the skill configuration, allowing new team members to inherit established processes without risk of accidental changes.

The feature is available to all Manus users as of February 12.

Natalie, a product marketing lead at Manus, demonstrated the feature in a post on X (video embedded below), showing how she configured a project with a curated set of skills for her team's launch video workflow:

Why this matters

The update addresses a practical challenge in team-based AI agent use: ensuring that different people running the same type of task get consistent results.

Project Skills is one approach to the problem. Rather than relying on each user to configure the agent correctly, teams can define the approved workflows once and enforce them across a project.

This is the second skills-related update Manus has shipped in roughly two weeks, suggesting active development in this area as the platform builds out team and enterprise features under Meta.