Files
codex/codex-rs/core/tests/suite/mod.rs
T
pakrym-oai 768848ab6f Add experimental turn additional context (#24154)
## Summary

Adds experimental `additionalContext` support to `turn/start` and
`turn/steer` so clients can provide ephemeral external context, such as
browser or automation state, without turning that plumbing into a
visible user prompt or triggering user-prompt lifecycle behavior.

## API Shape

The parameter shape is:

```ts
additionalContext?: Record<string, {
  value: string
  kind: "untrusted" | "application"
}> | null
```

Example:

```json
{
  "additionalContext": {
    "browser_info": {
      "value": "Active tab is CI failures.",
      "kind": "untrusted"
    },
    "automation_info": {
      "value": "CI rerun is in progress.",
      "kind": "application"
    }
  }
}
```

The keys are opaque and caller-defined.

## Context Injection

When provided, accepted entries are inserted into model context as
hidden contextual message items, not as visible thread user-message
items.

`kind: "untrusted"` entries are inserted with role `user`:

```text
<external_${key}>${value}</external_${key}>
```

`kind: "application"` entries are inserted with role `developer`:

```text
<${key}>${value}</${key}>
```

Values are not escaped. Each value is truncated to 1k approximate tokens
before wrapping.

For `turn/start`, accepted additional context is inserted before normal
user input. For `turn/steer`, additional context is merged only when the
steer includes non-empty user input; context-only steers still reject as
empty input.

## Dedupe Strategy

`AdditionalContextStore` lives on session state and stores the latest
complete additional-context map.

Each `turn/start` or non-empty `turn/steer` treats its
`additionalContext` as the current complete set of values. Entries are
injected only when the key is new or the exact entry for that key
changed, including `value` or `kind`. After merging, the store is
replaced with the provided map, so omitted keys are removed from the
retained set and can be injected again later if reintroduced.

Omitting `additionalContext`, passing `null`, or passing an empty object
resets the store to empty and injects nothing.

## What Changed

- Threads experimental v2 `additionalContext` through app-server into
core turn start and steer handling.
- Adds separate contextual fragment types for untrusted user-role
context and application developer-role context.
- Uses pending response input items so additional context can be
combined with normal user input without treating it as prompt text.
- Adds integration coverage for start/steer flow, role routing,
dedupe/reset behavior, deletion/re-add behavior, hook-blocked input
behavior, empty context-only steer rejection, external-fragment marker
matching, and truncation.
2026-05-26 13:02:34 -07:00

122 lines
3.1 KiB
Rust

// Aggregates all former standalone integration tests as modules.
use codex_apply_patch::CODEX_CORE_APPLY_PATCH_ARG1;
use codex_exec_server::CODEX_FS_HELPER_ARG1;
use codex_sandboxing::landlock::CODEX_LINUX_SANDBOX_ARG0;
use codex_test_binary_support::TestBinaryDispatchGuard;
use codex_test_binary_support::TestBinaryDispatchMode;
use codex_test_binary_support::configure_test_binary_dispatch;
use ctor::ctor;
// This code runs before any other tests are run.
// It allows the test binary to behave like codex and dispatch to apply_patch and codex-linux-sandbox
// based on the arg0.
// NOTE: this doesn't work on ARM
#[ctor]
pub static CODEX_ALIASES_TEMP_DIR: Option<TestBinaryDispatchGuard> = {
configure_test_binary_dispatch("codex-core-tests", |exe_name, argv1| {
if argv1 == Some(CODEX_CORE_APPLY_PATCH_ARG1) {
return TestBinaryDispatchMode::DispatchArg0Only;
}
if argv1 == Some(CODEX_FS_HELPER_ARG1) {
return TestBinaryDispatchMode::DispatchArg0Only;
}
if exe_name == CODEX_LINUX_SANDBOX_ARG0 {
return TestBinaryDispatchMode::DispatchArg0Only;
}
TestBinaryDispatchMode::InstallAliases
})
};
#[cfg(not(target_os = "windows"))]
mod abort_tasks;
mod additional_context;
mod agent_jobs;
mod agent_websocket;
mod agents_md;
mod apply_patch_cli;
#[cfg(not(target_os = "windows"))]
mod approvals;
mod cli_stream;
mod client;
mod client_websockets;
mod code_mode;
mod codex_delegate;
mod collaboration_instructions;
mod compact;
mod compact_remote;
mod compact_remote_parity;
mod compact_resume_fork;
mod deprecation_notice;
mod exec;
mod exec_policy;
mod fork_thread;
mod hierarchical_agents;
#[cfg(not(target_os = "windows"))]
mod hooks;
#[cfg(not(target_os = "windows"))]
mod hooks_mcp;
mod image_rollout;
mod items;
mod json_result;
mod live_cli;
mod mcp_turn_metadata;
mod model_overrides;
mod model_switching;
mod model_visible_layout;
mod models_cache_ttl;
mod models_etag_responses;
mod openai_file_mcp;
mod otel;
mod override_updates;
mod pending_input;
mod permissions_messages;
mod personality;
mod personality_migration;
mod plugins;
mod prompt_caching;
mod prompt_debug_tests;
mod quota_exceeded;
mod realtime_conversation;
mod remote_env;
mod remote_models;
mod request_compression;
#[cfg(not(target_os = "windows"))]
mod request_permissions;
#[cfg(not(target_os = "windows"))]
mod request_permissions_tool;
mod request_plugin_install;
mod request_user_input;
mod responses_api_proxy_headers;
mod resume;
mod resume_warning;
mod review;
mod rmcp_client;
mod rollout_list_find;
mod safety_check_downgrade;
mod search_tool;
mod shell_command;
mod shell_serialization;
mod shell_snapshot;
mod skill_approval;
mod skills;
mod spawn_agent_description;
mod sqlite_state;
mod stream_error_allows_next_turn;
mod stream_no_completed;
mod subagent_notifications;
mod tool_harness;
mod tool_parallelism;
mod tools;
mod truncation;
mod turn_state;
mod unified_exec;
mod unstable_features_warning;
mod user_notification;
mod user_shell_cmd;
mod view_image;
mod web_search;
mod websocket_fallback;
mod window_headers;
#[cfg(target_os = "windows")]
mod windows_sandbox;