Operational brief
OpenCode
Open-source coding agent for terminal, desktop and IDE sessions, with configurable tool permissions.
Choose it when
You want a session-led coding agent and are willing to match permission and plugin configuration to your chosen OpenCode generation.
Skip it when
You need a proven unattended service, or assume a local client means code never reaches a model provider or a sharing server.
01 / Classification
Autonomy evidence
This is our classification, not a vendor badge. The record says what the product normally does. The gaps stay visible.
02 / Operations
Control reality
Useful agents eventually touch something real. This is the operational picture we have, not a promise hidden in a sales page.
03 / Receipts
Evidence ledger
Here is what this brief is built on. “Documented” means we found a source. “Tested” means this desk actually ran it. Those are different things.
Sources on file
Corrections
- 2026-10-17 · Scoped release metadata to V1 v1.19.3, not a unified V1/V2 latest or an installed upgrade. Added V2's default outside-project access, custom-subagent permission boundary and missing LSP execution/diagnostics. Recast sharing claims as historical V1 documentation because the sharing-doc URL returned 404; current sharing/export remains unverified. Documented, not runtime-tested.
04 / Change log
Related coverage
The brief tells you what the product is. Payloads track the changes that may alter the decision.
No linked change coverage on file yet.