Packages

codex_sdk

0.4.2
0.18.1 0.18.0 0.17.0 0.16.1 0.16.0 0.14.0 0.13.0 0.12.0 0.11.0 0.10.1 0.10.0 0.9.0 0.8.0 0.7.2 0.7.1 0.7.0 0.6.0 0.5.0 0.4.5 0.4.4 0.4.3 0.4.2 0.4.1 0.4.0 0.3.0 0.2.5 0.2.4 0.2.3 0.2.2 0.2.1 0.2.0 0.1.0

Idiomatic Elixir SDK for OpenAI's Codex agent. Provides a complete, production-ready interface with streaming support, comprehensive event handling, and robust testing utilities.

HexDocs

Current section

Files

Jump to
Readme
32 Versions
6 Dependencies
Dependants
Files
Activity
codex_sdk docs 20251129 models models_compaction_and_prompts.prompt.md
Raw

docs/20251129/models/models_compaction_and_prompts.prompt.md 

# Prompt: Sync models, compaction, and prompt handling (TDD)



Required reading:

- docs/20251129/models/models_compaction_and_prompts.md

- docs/05-api-reference.md (model options, reasoning), docs/observability-runbook.md (compaction signals)

- lib/codex/items.ex, lib/codex/thread.ex, lib/codex/telemetry.ex (model selection, history, compaction events)

- test/codex/thread_test.exs, test/codex/items_test.exs, test/codex/telemetry_test.exs



Context to carry:

- Model defaults shift toward gpt-5.1 variants with reasoning-level tweaks; experimental tool-enabled models exist.

- Compaction is on by default; emits explicit compaction events and token-usage updates.

- Truncation helpers were refactored to avoid double truncation and align token accounting.



Instructions (TDD):

1) Read the docs to align on new defaults and compaction/truncation expectations.

2) Add failing tests for model default selection/reasoning options, compaction events, and token-usage updates.

3) Implement model list/default updates and compaction event handling; ensure truncation logic matches upstream (no double truncation).

4) Verify history replay/prompt sizing aligns with the refactored helpers; adjust tests accordingly.

5) Run targeted tests then `mix test`; keep scope to SDK model/prompt/compaction surfaces.