The AI Agent Index

Documenting the technical and safety features of deployed agentic AI systems

Devin


Basic information

Website: https://web.archive.org/web/20241217150025/https://www.cognition.ai/blog/introducing-devin

Short description: A general-purpose software engineering assistant.

Intended uses: What does the developer say it’s for? Rapid general-purpose coding and software development.

Date(s) deployed: Openly available [source]


Developer

Website: https://web.archive.org/web/20241224121630/https://www.cognition.ai/

Legal name: Cognition AI, Inc [source]

Entity type: Corporation [source]

Country (location of developer or first author’s first affiliation): Incorporation: Delaware, USA (COGNITION AI, INC. (2875752)) [source]. Business registration: California, USA Cognition AI, Inc. (6224891) [source]. HQ: New York [source]

Safety policies: What safety and/or responsibility policies are in place? Unknown


System components

Backend model: What model(s) are used to power the system? Variable, including OpenAI o1 [source]

Publicly available model specification: Is there formal documentation on the system’s intended uses and how it is designed to behave in them? None

Reasoning, planning, and memory implementation: How does the system ‘think’? Unknown

Observation space: What is the system able to observe while ‘thinking’? Devin operates in a workspace involving a browser, code editor, and shell and is able to observe the history of all [source].

Action space/tools: What direct actions can the system take? Writing and executing code and queries in its environment browser, editor, and shell [source].

User interface: How do users interact with the system? An application in which the user provide prompts; monitor the browser, code editor, and shell; and analyze outputs [source].

Development cost and compute: What is known about the development costs? Unknown


Guardrails and oversight

Accessibility of components:

  • Weights: Are model parameters available? N/A; backends various models
  • Data: Is data available? N/A; backends various models
  • Code: Is code available? Closed source
  • Scaffolding: Is system scaffolding available? Closed source
  • Documentation: Is documentation available? Available [source]

Controls and guardrails: What notable methods are used to protect against harmful actions? “For complex tasks, Devin offers to wait for your confirmation” [source]. Devin produces worklogs that users can view [source].

Customer and usage restrictions: Are there know-your-customer measures or other restrictions on customers? None

Monitoring and shutdown procedures: Are there any notable methods or protocols that allow for the system to be shut down if it is observed to behave harmfully? Unknown


Evaluation

Notable benchmark evaluations: 13.86% on SWE-Bench [source]

Bespoke testing: Demos [source] [source] [source].

Safety: Have safety evaluations been conducted by the developers? What were the results? None

Publicly reported external red-teaming or comparable auditing:

  • Personnel: Who were the red-teamers/auditors? None
  • Scope, scale, access, and methods: What access did red-teamers/auditors have and what actions did they take? None
  • Findings: What did the red-teamers/auditors conclude? None

Ecosystem information

Interoperability with other systems: What tools or integrations are available? Devin integrates with Slack, GitHub, and Linear [source] [source].

Usage statistics and patterns: Are there any notable observations about usage? Unknown


Additional notes

None