Agent Atlas / Coding agents

openinterpreter/openinterpreter

A Rust coding-agent fork that combines Codex lineage with local state, provider switching, and selectable harness behavior aimed at lower-cost models.

Coding agent Full dossier Reported + sourced Reviewed July 21, 2026
Intended operator

Developers who want a local agent and interchangeable model or harness behavior.

Primary surfaces

terminal agent / local runtime / provider and harness selection

Tracked repository

openinterpreter/openinterpreter

Teaching mechanism

A maintained fork inherits leverage and reconciliation work

The project repeatedly merges Codex upstream while owning release, installer, cache, documentation, and harness changes for a different product bet.

Evidence limitPublic fork lineage proves technical inheritance, not privileged coordination, adoption, or production reliability.
Inspect the source

System Map

Eight inspection boundaries
Model boundary

Provider switching and selectable harnesses adapt a Codex-derived runtime to different model behavior.

Context and memory

The project emphasizes local state while inheriting and modifying upstream session machinery.

Tools and execution

Codex-derived execution and tool paths provide the base for fork-owned harness behavior.

Permissions and isolation

Inherited safety machinery still has to be reconciled with fork-owned changes and releases.

Sessions and recovery

Local state, installer behavior, and upstream session changes create a continuing compatibility obligation.

Delegation

Delegation capability follows the evolving upstream base and requires fork-specific verification.

Human control

The terminal product and harness selection expose the fork's operator-facing choices.

Evidence surface

Merge ancestry, release workflows, installer code, docs, and fork-owned tests distinguish inheritance from new work.

Learn Through This Project

8 lessons
Chapter 1 Before the First Model Request Chapter 2 The Invisible Prompt Chapter 3 Tool Calls: Contract, Authorization, Execution Chapter 4 Loops, Workflows, and Recovery Chapter 7 The Approval Trap Chapter 8 The Human Control Loop Chapter 9 Trust Must Make Its Case Chapter 10 The Future Has Three Branches

Compare Its Boundaries

7 labs
Comparison labTool ContractsComparison labGoverned MemoryComparison labApproval BoundariesComparison labHuman WorkbenchComparison labSessions And RecoveryComparison labEvidence And EvaluationComparison labMaturity Signals

Latest Reporting

1 article
Open Interpreter: roots, maintainers, and the contributor advantage

July 16, 2026 — The new Rust Open Interpreter is openly building on Codex; its public advantage is upstream leverage, not a proven inside track. openinterpreter/openinterpreter, openai/codex

Primary Sources

Open Interpreter repository

Primary current Rust implementation and fork history.

Open Questions

  • Can the fork keep absorbing upstream safety and protocol changes while proving distinct harness behavior?

Evidence state describes what this Atlas profile can support today. It is not a product score and it is not evidence that uninspected capabilities are absent.