## What changed
Configure managed memory consolidation agents with a `PermissionProfile` directly and explicitly bind both configuration and permission workspace roots to the memory directory. Keep network access disabled and exclude temporary directories from writable roots. Preserve disabled and externally managed parent permission profiles.
## Testing
Extend regression coverage to check that inherited workspace roots remain readable but are not writable, protected metadata stays read-only whether or not it exists, and incompatible permission constraints prevent agent configuration. Verify effective filesystem enforcement, network policy, and disabled approvals across parent permission profiles.
GitOrigin-RevId: b1ae115cade059da9e5fb83c218a1d89185b4a70
## Why
Memory consolidation inherits the session's service tier. Give background consolidation its own setting, defaulting to Standard.
## What changed
Add `memories.consolidation_service_tier` with a default of `"default"` and apply it to the consolidation agent. Normalize recognized tier aliases, including `"fast"` to `"priority"`, and expose the setting in the configuration schema.
## Testing
Add coverage for configuration parsing and alias normalization, Standard consolidation despite an Ultrafast session tier, and an explicit `"flex"` consolidation override.
GitOrigin-RevId: 58031ad56196152d212531d56e7fe9813f83317e
## Why
Concurrent imports and consolidation can modify the same memory workspace, leaving a partial import across memory versions or interfering with consolidation snapshots and baselines.
## What changed
- Add a cross-process workspace lock shared by memory imports and consolidation.
- Acquire locks for every import destination before modifying any workspace. Return a retryable busy error if a destination is locked, and skip consolidation when its workspace is busy.
- Hold locks through import completion even if the caller drops the future, and through consolidation agent termination and baseline finalization.
## Testing
Add an app-server integration test that blocks an active consolidation, verifies import rejection without partial writes or snapshot changes, then retries successfully and confirms imported resources remain visible to the next consolidation.
GitOrigin-RevId: c4ce945f916dcb4d75a3190ea869751fac51de83
## What changed
Emit a `codex.memory.storage_bytes` histogram after successful phase-two consolidation, including runs with no workspace changes. Sum regular-file sizes recursively from filesystem metadata, excluding `.git` entries and symbolic links. Stop the phase-two duration timer before measuring storage.
## Testing
Add coverage for both memory versions, nested files, UTF-8 byte counts, Git metadata exclusion, and Unix symbolic links, including broken links and directory loops. Extend the workspace baseline reset test to verify the remaining storage size.
GitOrigin-RevId: 9170c34d656728897790decd2e2cabb38a81836f
## What changed
- Consolidate v2 rollout summaries into `memory_summary.md` without generating `raw_memories.md` or requiring `MEMORY.md`. Validate the summary's required sections and size below 10,000 UTF-8 bytes.
- Add v2 read instructions for selective history retrieval, evidence-grounded preferences, citations, and explicit memory edits. Split injected instructions into fragments to preserve the complete summary within fragment byte limits.
- Record memory citation usage in the store selected by `memories.version`.
## Testing
Add coverage for v2 consolidation without a handbook, summary validation, version-isolated memory reads, and resetting both memory versions while preserving threads.
GitOrigin-RevId: 1d895fb4a23a973f1a45ba03be07a1f480c10227
## What changed
- Add `memories.version` with `v1` as the default and `v2` as an opt-in selection.
- Route memory generation, summaries, and retrieval tools through the selected version. Store `v2` artifacts in `memories_v2` and lazily create a separate SQLite database, keeping jobs and outputs isolated while sharing the thread catalog.
- Preserve the initial memory version across extension config updates so summaries and retrieval tools use the same namespace.
- Clear both versions on memory reset and remove thread memory from both stores on thread deletion. Recognize both artifact roots in shell usage telemetry.
## Testing
Add regression coverage for version defaults and validation, independent job claims and outputs, deletion and reset across versions, and shell usage classification for both memory roots.
GitOrigin-RevId: aa799bf87ab6ec10c8f5668213931e57547d080b
## Why
Background memory consolidation must honor managed completion policies without
running project completion checks or sending user-facing notifications.
## What changed
- Add a memory-consolidation stop target that selects managed policy hooks and
executor-scoped cleanup while excluding user, project, session, and plugin
stop hooks.
- End memory consolidation with an error when a managed stop hook rejects it,
and propagate that turn error into agent status.
- Disable legacy completion notifications for the background memory worker.
## Testing
- Cover hook selection, executor cleanup, managed rejection, agent error
status, and notification suppression.
GitOrigin-RevId: be350679ddcde2dde8a408b0972f08facff01a37
## What changed
- Reject a symbolic link used as the memory root.
- Recursively remove symbolic links before preparing or seeding the memory workspace.
- Remove and reject symbolic links created during consolidation, including when the worker fails, so they cannot affect files outside the workspace.
## Testing
Add Unix startup tests covering pre-existing extension links and links created by both successful and failed consolidation workers.
GitOrigin-RevId: f96e03812773f7beaa16455dd77370ee47d8a71f
## Why
A phase-two memory job could release its lease while its consolidation agent was
still shutting down, allowing another worker to race the existing agent.
## What changed
- Await consolidation-agent shutdown before validating artifacts and completing
the job.
- Keep the lease until it expires if shutdown fails, and remove the agent from
the thread manager only after shutdown succeeds.
## Testing
Update startup tests to wait for the phase-two job lifecycle to finish and verify
that the consolidation thread has been removed.
GitOrigin-RevId: 7ac3a7369529822ac95fd2ee9dcb27d39c191d20
## What changed
- Remove the `spawn_agents_on_csv` and `report_agent_job_result` tools and their agent-job runtime and state models.
- Drop the legacy `agent_jobs` and `agent_job_items` tables during state database migration.
- Keep `features.enable_fanout` and `agents.job_max_runtime_seconds` accepted as no-op compatibility settings while omitting them from the generated configuration schema.
## Testing
- Verify upgrades remove both legacy agent-job tables.
- Verify the removed feature and configuration keys still parse without taking effect.
GitOrigin-RevId: 8cc3337da78c67162229f02f40a747f503542646
## What changed
- Parse `cache_write_tokens` from response input token details and carry it through token usage aggregation.
- Expose `cache_write_input_tokens` in protocol, app-server, exec, and TypeScript SDK usage events, defaulting it to zero for compatibility with older payloads.
- Report cache-write usage in analytics, metrics, traces, and rollout data alongside existing cache-read usage.
## Testing
- Add coverage for response parsing, usage event output, aggregation, analytics, metrics, and telemetry fields.
GitOrigin-RevId: 634e4fbe3086a8923c0ee13b622573e7d9d221bd
## What changed
- Pass the parent turn's effective permission profile to the memory consolidation agent, including thread-level permission and legacy sandbox overrides.
- Preserve disabled and externally enforced permission profiles instead of replacing them with a managed sandbox.
- Continue restricting consolidation to the memory root without network access when the parent uses Codex-managed permissions.
## Testing
- Add coverage for disabled, external, and managed parent permission profiles.
GitOrigin-RevId: 9ca3be0e41dc14d858053f6f70e33f4ae7c578e1
## Why
The memory consolidation agent changes its working directory to the memory
root, but applying its sandbox policy directly to the permissions object can
leave workspace roots inherited from the parent configuration.
## What changed
Apply the consolidation sandbox policy through `Config` so its workspace roots
are synchronized with the memory root. Add a test covering the working
directory, workspace roots, and effective legacy sandbox policy.
GitOrigin-RevId: dc6e75aef1ffa84a2f66fec6ed3a6fedcf9aa672
## Why
A completed Phase 2 agent run does not guarantee that its required outputs were
created correctly. Treating a clean workspace as success can also preserve a
state where those outputs are missing.
## What changed
- Require `MEMORY.md` to be a file and `memory_summary.md` to start with `v1`
before marking consolidation successful.
- Fail completed runs with invalid artifacts without resetting the workspace
baseline, allowing the job to be retried.
- Run consolidation for a clean workspace when its required artifacts are
invalid instead of taking the no-change success path.
## Testing
Added coverage for rejecting an invalid summary and retrying a clean workspace
whose consolidation artifacts are missing.
GitOrigin-RevId: ac57b2ba9d062c5203ca51afd716795a91814bb3
## Why
Memory startup used hardcoded OpenAI model slugs for extraction and
consolidation. That works for the default OpenAI-compatible path, but
provider-specific backends can require different model identifiers. In
particular, Amazon Bedrock should use its Bedrock model ID for these
background memory requests instead of the OpenAI `gpt-5.4-mini` /
`gpt-5.4` slugs.
## What Changed
- Added provider-owned preferred memory model methods alongside
`approval_review_preferred_model`.
- Updated memory extraction and consolidation to resolve their default
model through the active `ModelProvider`.
- Added Amazon Bedrock overrides so both memory stages use
`openai.gpt-5.4` through Bedrock’s provider-specific model ID.
- Kept explicit `memories.extract_model` and
`memories.consolidation_model` config overrides taking precedence.
- Added startup coverage for default OpenAI and Bedrock memory model
selection.
#closes #26288
## Summary
Generated memory rows and their stage-one/stage-two job state currently
live in `state_5.sqlite` alongside thread metadata. That makes memory
cleanup and regeneration share the main state schema even though those
rows are memory-pipeline data and can be rebuilt independently from the
durable thread records.
This PR moves the memory-owned tables into a dedicated
`memories_1.sqlite` runtime database while keeping thread metadata in
`state_5.sqlite`.
## Changes
- Adds a separate memories DB runtime, migrator, path helpers, telemetry
kind, and Bazel compile data for `state/memory_migrations`.
- Introduces `MemoryStore` behind `StateRuntime::memories()` and moves
memory table/job operations onto that store.
- Drops the old memory tables from the state DB and recreates their
schema in `state/memory_migrations/0001_memories.sql`.
- Updates memory startup, citation usage tracking, rollout pollution
handling, `debug clear-memories`, and app-server `memory/reset` to
operate through the memories DB.
- Preserves cross-DB behavior by hydrating thread metadata from the
state DB when selecting visible memory outputs and checking stage-one
staleness.
## Verification
- Added/updated `codex-state` tests for deleted-thread memory visibility
and already-polluted phase-two enqueue behavior.
- Updated `debug clear-memories`, app-server `memory/reset`, and
memories startup tests to seed and assert memory rows through
`memories_1.sqlite`.
## Why
Memory startup was tied to thread lifecycle events such as create, load,
and fork. That can run memory work before a thread receives real user
input, and it makes startup cost scale with thread management instead of
actual turns. Moving the trigger to `thread/sendInput` keeps memory
startup aligned with the first real user turn and lets it use the
current thread config at turn time.
The idea is to prevent ghost cost due to pre-warm triggered by the app
Turn-based startup can also make global phase-2 consolidation easier to
request repeatedly, so this adds a success cooldown and tightens the
default startup scan window.
## What Changed
- Start `codex_memories_write::start_memories_startup_task` after a
non-empty `thread/sendInput` turn is submitted, instead of from thread
create/load/fork paths:
https://github.com/openai/codex/blob/d4a6885b7829e2fd2ec7a09355e4f75ebe1d1fe3/codex-rs/app-server/src/codex_message_processor.rs#L6477-L6487
- Expose `CodexThread::config()` so app-server can pass the live config
into memory startup at turn time.
- Add a six-hour successful-run cooldown for global phase-2
consolidation via `SkippedCooldown`:
https://github.com/openai/codex/blob/d4a6885b7829e2fd2ec7a09355e4f75ebe1d1fe3/codex-rs/state/src/runtime/memories.rs#L963-L966
- Reduce memory startup defaults to at most 2 rollouts over 10 days:
https://github.com/openai/codex/blob/d4a6885b7829e2fd2ec7a09355e4f75ebe1d1fe3/codex-rs/config/src/types.rs#L31-L34
## Verification
Updated the memory runtime coverage around phase-2 reclaim behavior,
including `phase2_global_lock_respects_success_cooldown`.
---------
Co-authored-by: Codex <noreply@openai.com>
Keep extracting memories out of core and moving the write trigger in the
app-server
This is temporary and it should move at the client level as a follow-up
This makes core fully independant from `codex-memories-write`
---------
Co-authored-by: Codex <noreply@openai.com>
ioanbudea-OAI