SPIRITT rises to 20/36 as a Domain Specialist
What changed: SPIRITT moved from 14 to 20/36 and is now classified as a Domain Specialist (L3) following a submission-triggered review. OxCode also rose from 14 to 19/36, while Amelia moved from 9 to 15/36.
Why it matters: Operators choosing a specialized tool now have two closely scored Domain Specialists to assess, rather than treating a single score change as a standalone signal. Amelia’s move places it at 15/36 as an Adaptive Collaborator (L2). The scores do not by themselves establish workflow fit.
Do this: Compare SPIRITT’s 20/36 listing with OxCode’s 19/36 listing against one domain-specific workflow before deciding which deserves a pilot.
See what changed on SPIRITT's listing →
What changed on the tools' own sites
|
Eightfold AI introduces AI Recruiting Teammate for autonomous recruiting… Eightfold AI added an AI Recruiting Teammate that coordinates agents, tools, and hiring workflows, including AI Interviewer.
Matters if you coordinate hiring workflows across recruiting teams.
|
|
Revdoku adds AI-powered support tickets and agent-managed docs workflows Revdoku added page-level AI assistance, ticket creation from unanswered questions and email, drafted replies, and authenticated document management access.
Matters if you maintain documentation and handle support requests.
|
|
CrePal launches an AI Video Creation Agent with conversational editing… CrePal added an AI video creation agent that plans productions, coordinates image and video models, and supports conversational revisions.
Matters if you produce videos from ideas, scripts, or PDFs.
|
|
ChatterMate expands its open-source AI support agent with in-app actions… ChatterMate expanded its support agents with approved in-app actions, user guidance, and ticket investigation using logs and databases.
Matters if you operate customer support agents inside your application.
|
|
Coarena adds agent workflows and MCP access alongside revised free API limits Coarena added computer and Android phone agent workflows, plus MCP access to its API verbs.
Matters if you build agents that operate computers or Android phones.
|
Changes we spotted on each tool's own website since the last issue.
Score moves
Source: agentic.ai's rubric re-evaluations.
|