updated 2026-10-09 20:07 UTC

Codex

Repository by openai

Lightweight coding agent that runs in your terminal

Coding Agents Rust Listed Apache-2.0

Is Codex safe? Strong public trust posture, backed by multiple independent signals.

OSSF Scorecard 6.0 / 10
Provenance Attested
Signed commits 0%
Last push today
Advisories None known
Compare Codex

How does it stack up against its Coding Agents neighbours?

Pick any agent to compare →
Suggest a correction

In detail: Codex scores 90.0/100 (Grade A), ranked #20 of 1371 tracked open-source AI agent projects, on evidence coverage B (3 of 5 independent signal types). The public evidence: its packages ship with cryptographic provenance; OSSF Scorecard rates its supply-chain practices 6.0/10; recent commits are unsigned; no published advisory affects its latest release (@openai/codex, openai-codex); last pushed 2026-10-09. Every point is earned from checkable signals — never paid placement. How scoring works →

What Codex can reach, and who runs it

Checked 2026-10-06, each with its source. PUBLISHER is the company's or project's own statement, which we haven't verified; CHECKED is something we observed ourselves. These facts describe the product. They don't change its score.

Who runs it

  • CHECKED Made by OpenAI. The CLI is open source under Apache-2.0. github.com

What it can reach

  • PUBLISHER In a version-controlled folder it recommends its Auto preset, which reads, edits and runs commands in the working directory without asking. Writes stay in that workspace, and network access is off unless you enable it. learn.chatgpt.com
  • PUBLISHER Those limits are enforced by the operating system: Seatbelt on macOS, bwrap with seccomp on Linux, and a WSL2 or native sandbox on Windows. learn.chatgpt.com

Security practice

  • CHECKED openai.com's security.txt points to a Bugcrowd program, and the repository has a SECURITY.md. openai.com
  • CHECKED Its maintainers have published 1 security advisory (high), in September 2025. github.com
Something wrong or out of date? Suggest a correction

How Codex could raise its score

Each line changes one public signal and recomputes with the live scoring function. The gains don't add up exactly, because the score is capped near the top.

Already done? Suggest a correctionAdd the trust badge

Chain of custody

Scanners check what the code says. This traces who ships Codex and whether that has changed: from the source repository, through how changes are reviewed and released, to the package you install. These are the checks for OWASP ASI04 Agentic Supply Chain Vulnerabilities.

  1. Source Unsigned

    github.com/openai/codex, Apache-2.0, last pushed 2026-10-09. None of the last 100 commits carry a verified signature, so no change traces to a verified identity.

  2. Review and release Mixed

    OpenSSF Scorecard rates the repository's practices 6.0/10 (scanned Oct 8, 2026). The checks that decide who can get a change released:

    • Code Review0
    • Branch Protection4
    • Signed Releases8
    • Dangerous Workflow10
    • Token Permissions0
    • Pinned Dependencies9
    All 18 Scorecard checks
    Binary-Artifacts 10
    Branch-Protection 4
    CI-Tests 10
    CII-Best-Practices 0
    Code-Review 0
    Contributors 10
    Dangerous-Workflow 10
    Dependency-Update-Tool 10
    Fuzzing 0
    License 10
    Maintained 10
    Packaging 10
    Pinned-Dependencies 9
    SAST 0
    Security-Policy 10
    Signed-Releases 8
    Token-Permissions 0
    Vulnerabilities 0

  3. Published packages Linked and attested

    • npm @openai/codex Source link points back to this repo Build provenance attested
    • PyPI openai-codex Source link points back to this repo Build provenance attested

  4. What you install today No known advisories

    Latest releases checked: @openai/codex 0.162.0, openai-codex 0.162.0. No published advisory affects it (OSV, checked 2026-10-09).

Changes to this chain

No change to package provenance, package source links, Scorecard coverage or license in HVTracker's daily snapshots of Codex.

Evidence behind this score

Coverage B: 3 of 5 independent evidence types found.

  • GitHub repository data
  • Package downloads
  • Supply-chain checks
  • Public actions (missing)
  • Community mentions (missing)

Verify this score yourself

Every build re-issues Codex’s score as an Ed25519-signed credential, valid for 7 days. You can check it offline against HVTracker’s published key; if anyone changes a number after signing, verification fails.

Signed credential (JSON) How to verify

What this score doesn’t check

It covers who ships the code, not what the code does. It doesn’t read tool descriptions or prompts for injected instructions, watch runtime behaviour, or find bugs nobody has disclosed yet. For that, run a content scanner before you connect it, such as Cisco MCP Scanner or Snyk Agent Scan.

How Codex compares in Coding Agents

  1. #1 Cline 91.7 +1.7
  2. #2 Codex 90.0 this agent
  3. #3 Qwen Code 89.4 −0.6
  4. #4 Bernstein 88.7 −1.3
  5. #5 ECC 87.8 −2.2

Bars show each HVTrust score; the tick marks Codex’s 90.0.

Where the 90.0 comes from

HVTrust dimensions vs the Coding Agents average

90.0 / 100 · 100.0% confidence

Codex Coding Agents average (77 agents)

Safety / Integrity50% OSSF Scorecard · 30% provenance · 20% signed commits
15.0 / 25
6.0 above avg 9.0
Identity / Provenance60% listing status · 40% build provenance
18.0 / 18
6.3 above avg 11.7
Transparency50% declared license · 50% OSSF Scorecard
13.6 / 17
1.8 above avg 11.8
Maintenance60% last-push freshness · 40% commit activity
20.0 / 20
5.0 above avg 15.0
AdoptionLog-scaled stars · package downloads
20.0 / 20
7.9 above avg 12.1

Quick Trust Read

What Would Improve It
Increase the share of verified signed commits for clearer maintainer identity.
Maintainer Checklist
Raise Scorecard signals Current OSSF Scorecard is 6.0/10. Tighten the weakest checks to improve public safety evidence.
Increase signed commits Raise the share of verified-signed commits to make maintainer identity and release history easier to trust.
Compare with Cline Get email alerts for Codex View Coding Agents Report missing data
100.0
Activity sub-score · out of 100
#2
In Coding Agents

How to read this: HVTrust (0–100) weighs supply-chain signals (provenance, OSSF Scorecard, signed commits, open license) alongside real-world adoption. Grade A reflects the trust score band: A ≥ 80, B ≥ 65, C ≥ 50, D < 50. Evidence coverage B is separate — it grades how many independent signal types back the score (3 of 5), so a high score on thin evidence stays visible. Full methodology →

Signals refreshed 2026-10-09 20:07 UTC · Repo last pushed today

Rank Trend

2026-09-29 2026-10-09

Activity & Reach

Stars
128.4k
Forks
20.1k
Last Push
2026-10-09
today
Commits (4 wk)
1548
Downloads (7d)
21,862,626
npm+pypi
HN mentions (30d)
—
Open Issues
21708
Rank Change
▲2
was #22

Analysis

Activity Inputs

100.0 / 100
StarsRepository reach
30 / 30
FreshnessLast push recency
25.0 / 25
ActivityRecent commits
25 / 25
CommunityFork signal
20 / 20

Common questions about Codex

Public supply-chain signals for Codex are strong: it has multiple independent trust indicators in place. This does not replace your own security review, but Codex carries less obvious unverified-evidence risk than projects with thin signals.
Does Codex publish package provenance?
Yes. Codex's package releases carry build provenance attestations, which cryptographically link the published package back to its source repository and CI workflow.
Does Codex have an OpenSSF Scorecard?
Codex has an OpenSSF Scorecard score of 6.0/10. The Scorecard checks for branch protection, signed releases, dependency updates, fuzzing, code review, and other supply-chain hygiene items. See the full check breakdown on this page.
Is Codex actively maintained?
Actively maintained. The repository was pushed to within the last 1 day(s).
What license does Codex use?
Codex ships under Apache-2.0. A declared, OSI-approved license is one of the transparency signals HVTrust scores.
Are Codex's commits signed?
0% of the last 100 commits to Codex are verified-signed (GPG, SSH, S/MIME, or GitHub's signing flow). Signed commits help confirm that code was authored by who the commit claims.

Not a safety endorsement. HVTracker describes what public signals show, not whether a project is safe for your use case. Run your own security review before adopting in production.

Compare Codex head-to-head

Codex vs Cline → Codex vs Qwen Code → Codex vs Bernstein →

AI agent surface

MCP, providers, tool surface
Scored in HVTrust

These runtime-trust fields — detected from public repo docs and manifests — contribute a bounded adjustment to this project's HVTrust score alongside supply-chain evidence. The exact values each field can add or subtract are documented in the methodology → Compare this surface across every listed agent in the capability matrix →

MCP Server Support
high confidence
Implemented
Codex appears to expose MCP server capabilities.
Detailed evidence is not shown in the public view.
External Service Dependencies
high confidence
1 detected
Public provider/service dependencies detected.
Credential signal: No explicit API-key/config marker detected.
Tool / Plugin Surface
medium confidence
Extensions
Extension based plugin/integration surface detected.
Detailed evidence is not shown in the public view.
Package Provenance Drift
high confidence
Match
Published package metadata matches the tracked repo
Detailed evidence is not shown in the public view.
  • MCP signal live
  • External deps live
  • Tool / plugin surface live
  • Package provenance drift live
How this surface has changed

Detected changes to Codex's runtime surface and supply-chain posture, from daily public-signal snapshots. A change here means our detectors see something different — a genuinely changed capability, or better evidence of an existing one.

2026-09-03
Tool Surface Changed
Detected tool/plugin surface changed: declared → extension-based

Maintain Codex?

For maintainers

HVTrust scores Codex from public signals only — we never contact maintainers first. If a signal is wrong, stale, or missing (provenance you publish, a Scorecard you run, signed releases), tell us and we'll review it. Corrections are public and tracked on GitHub.

Submit a correction → Add the trust badge

Reputation Timeline

Signal history
HVTrust 1Surface 1
2026-09-03
Tool Surface Changed
Detected tool/plugin surface changed: declared → extension-based
2026-09-03
HVTrust Changed
HVTrust down 3.8pts (93.8 → 90.0)
GitHub → PyPI: openai-codex npm: @openai/codex

Embed Badge Badge guide for maintainers →

For maintainers
HVTrust 90.0 Grade A
Markdown:
[![HVTrust](https://hvtracker.net/badge/codex.svg)](https://hvtracker.net/agents/codex)
HTML:
<a href="https://hvtracker.net/agents/codex"><img src="https://hvtracker.net/badge/codex.svg" alt="HVTrust"></a>

Other agents in Coding Agents

Data sources
GitHub REST API (repo, commits, stars, forks, license) · npm Registry (downloads, provenance) · PyPI / pypistats (downloads, provenance) · OpenSSF Scorecard CLI
Each agent's signals refresh once daily across 6 staggered batches. Methodology v4.4 · Raw JSON