Skip to main content

To-do Fixer

to-do-fixer executes directory-backed tasks in fresh Codex sessions. It selects pending files randomly and processes them one at a time until the directory is empty or Codex reports a global usage or authentication limit.

Requirements

  • Python 3.10 or newer
  • Git
  • An installed and authenticated codex CLI
  • Local GitHub repository clones beneath the selected repository root

Installation

From this directory:

python3 -m pip install .

Task format

Each non-hidden regular file directly inside the task directory is a pending task. The first line is the GitHub repository in OWNER/REPO format. Every remaining byte of UTF-8 text is used as the Codex prompt.

kaizten/example-repository
Implement the requested feature, run its tests, and update the documentation.

Repository clones are identified from their GitHub origin remote rather than from their directory name. A missing clone or multiple matching clones causes that task to fail.

Usage

to-do-fixer

Tasks are read from ~/.to-do-fixer/tasks and repositories are searched recursively beneath ~ by default. Select other locations with:

to-do-fixer --tasks-directory ~/codex-tasks --repos-root ~/workspaces

Codex runs sequentially with the workspace-write sandbox and --approve-for-me. A structured final response is required before a task is considered complete.

Completed tasks move into done/. Invalid, missing, ambiguous, blocked, or otherwise failed tasks move into failed/. Existing files in either directory are never overwritten. If Codex reaches a usage, quota, rate, credit, or authentication limit, the current task remains pending and the command stops.

Only one to-do-fixer process may use a task directory at a time. Pressing Ctrl-C stops the active Codex process and leaves its task pending.

Exit codes

Code Meaning
0 Every pending task was processed. Some may have moved to failed/.
1 Startup or configuration failed.
2 A global Codex limit or authentication failure stopped processing.
130 The user interrupted processing.

Development

python3 -m unittest discover -s tests -v
./build-package.sh

Metadata

Release files for to-do-fixer 0.0.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for to-do-fixer 0.0.2
File Size Uploaded Provenance Actions
to_do_fixer-0.0.2.tar.gz 9.0 kB Without provenance Details for to_do_fixer-0.0.2.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for to-do-fixer 0.0.2
File Interpreter ABI Platform Actions
to_do_fixer-0.0.2-py3-none-any.whl Python 3 none any Details for to_do_fixer-0.0.2-py3-none-any.whl

Total release size: 15.8 kB

Release files / to_do_fixer-0.0.2.tar.gz

Download URL to_do_fixer-0.0.2.tar.gz
Size 9.0 kB
Tags Source
SHA-256 checksum
How to use checksums
c60220b9ebf3754118fcf4bfe1fc2d742ea00c932861749947c05b00e8ceb15c
BLAKE2b-256 checksum
How to use checksums
159cf3b49c911ba68ec7981e5f7bc322e81f53bb59a44d1f8f816e92d10090c8
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.14.4

Release files / to_do_fixer-0.0.2-py3-none-any.whl

Download URL to_do_fixer-0.0.2-py3-none-any.whl
Size 6.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
732c536f96c6e68477912ecd379122dcfadc3fc0e0df9e6b63ce24bb00660b4c
BLAKE2b-256 checksum
How to use checksums
2c9212795aff033a443d99c2afc81dd2327c9afdf24366d7b9f585bdbdc86309
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.14.4

Security for to-do-fixer 0.0.2

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page