relay-codex 100/100 — SaferSkills

Is relay-codex safe to install?

SaferSkills independently audited relay-codex (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

100

35.0 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 0 flagged

Securityscore 100 · 0 findings

✓—All security checks passedNo findings in this category for the latest scan.pass

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

skills/relay-codex/SKILL.md · 1 file

skills/relay-codex/SKILL.md2.1 KB · Markdown

name: relay-codex

description: Use when another local agent needs a one-shot Codex CLI answer and interactive behavior is acceptable or controlled.

Purpose

Send a bounded prompt to the local Codex CLI through Agent Powerups relay when Codex is useful as an external reviewer or implementation advisor.

Compatible with: claude-code, gemini-cli, generic

Requires:

Local Codex CLI (codex)

When to Use

You want Codex to review a plan, diff, or small implementation question.
A one-shot answer is enough and persistent context is not needed.
The user has explicitly accepted local Codex CLI use.

Do not use when an interactive TUI would block automation. Check local Codex behavior first.

Inputs

Relay session name
Focused prompt with repo context, files, constraints, and output format
Optional note about whether Codex may suggest edits or only review

Workflow

Verify the CLI is available and note whether it opens an interactive UI:

codex --version

Start a Codex relay session:

apx relay start <name> --provider codex

Send one focused prompt:

apx relay ask <name> "<prompt>"

Read the relay artifact and verify claims locally before acting.
Stop the session unless it is deliberately needed:

apx relay stop <name>

Output

A relay artifact containing Codex's response
A decision on whether the response is accepted, rejected, or needs another review
Follow-up prompts only when they add new evidence

Verification

[ ] codex --version worked or missing CLI was reported.
[ ] Prompt stated whether the task was review-only or implementation advice.
[ ] Output artifact was read and checked against source files.
[ ] Any suggested command or edit was independently verified.
[ ] Session lifecycle was handled explicitly.

Failure Modes

Assuming non-interactive behavior across Codex versions.
Using unsafe automation flags as a default.
Asking broad implementation questions with no file scope.
Treating Codex output as approval to skip local tests.

relay-codex— agent skill

Is relay-codex safe to install?

Purpose

When to Use

Inputs

Workflow

Output

Verification

Failure Modes

Audit the pieces. Scan the whole. Decide.