The AI Agent Index

Documenting the technical and safety features of deployed agentic AI systems

Code Droid


Basic information

Website: https://web.archive.org/web/20240905220438/https://www.factory.ai/droids

Short description: Agent for software development [source]

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

Date(s) deployed: Code Droid documentation was posted June 18, 2024


Developer

Website: https://web.archive.org/web/20241224152919/https://www.factory.ai/

Legal name: The San Francisco AI Factory Inc [source]

Entity type: Corporation [source]

Country (location of developer or first author’s first affiliation): USA [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 models from Anthropic and OpenAI [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’? Only high level details are given. Droid can decompose tasks, “simulate decisions, perform self-criticism, and reflect on real and imagined decisions”. Droid uses “HyperCode to construct a multi-resolution representation of a given engineering system” and ByteRank for retrieving task-relevant information [source].

Observation space: What is the system able to observe while ‘thinking’? From the codebase, Droid can “autonomously construct explicit (graph) and implicit (latent space similarity) relationships within low-level data and extract insights about these relationships at different levels of abstraction” [source].

Action space/tools: What direct actions can the system take? Droid has access to developer tools and environments (intended to be similar to what is available to human developers) [source].

User interface: How do users interact with the system? Unknown

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? Unavailable, but they have a technical report [source]

Controls and guardrails: What notable methods are used to protect against harmful actions? 1. “Code Droid operates within a strictly defined, sandboxed environment that isolates its operational scope from main development environments.” 2. “Droids log and report the reasoning behind all of their actions.” 3. “DroidShield performs real-time static code analysis to detect potential security vulnerabilities, bugs, or intellectual property breaches before they are committed to code” [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: 19.27% on SWE-bench Full and 31.67% on SWE-bench Lite [source]

Bespoke testing: None

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? None

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


Additional notes

None