codex
Hand off a task to Codex CLI for autonomous execution. Use when a task would benefit from a capable subagent to implement, fix, investigate, or review code. Codex has full codebase access and can make changes.
83
Quality
76%
Does it follow best practices?
Impact
99%
2.67xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./data/skills-md/0xbigboss/claude-code/codex/SKILL.mdEvaluation results
100%
46%Add Email Validation to a Standalone Script
Codex command construction and git-repo detection
--skip-git-repo-check present
100%
100%
--full-auto included
100%
100%
--model omitted
100%
100%
--json flag present
0%
100%
-o flag with summary path
0%
100%
Progress JSONL redirection
0%
100%
mkdir -p before exec
0%
100%
CTCO structure used
0%
100%
Rationale: --model omission explained
100%
100%
Rationale: --skip-git-repo-check explained
100%
100%
Without context: $0.3249 · 1m 45s · 16 turns · 17 in / 5,357 out tokens
With context: $0.2645 · 1m 6s · 12 turns · 259 in / 3,783 out tokens
97%
53%Review a Large Codebase for Security Issues
Background execution and sandbox mode selection
--sandbox read-only used
0%
100%
--full-auto absent
100%
100%
Background execution noted
100%
100%
Do not check output immediately
20%
70%
wc -l for progress check
40%
100%
tail -n 3 for status
0%
100%
Summary file existence check
62%
100%
No full progress file read
75%
100%
--json and -o flags present
33%
100%
Session output directory used
0%
100%
Without context: $0.4285 · 2m 4s · 20 turns · 25 in / 6,829 out tokens
With context: $0.5021 · 2m 4s · 22 turns · 25 in / 6,811 out tokens
100%
85%Investigate Flaky Integration Tests
Context depth assessment and result reporting
Full context classification
25%
100%
Git state commands present
30%
100%
Medium/full context git commands
12%
100%
Session context summarized
80%
100%
CTCO structure used
0%
100%
Constraints section content
0%
100%
Output section with 5 bullets
0%
100%
Result uses ## Codex Result format
10%
100%
--full-auto flag present
0%
100%
--json and -o flags in command
0%
100%
mkdir -p before codex exec
0%
100%
Without context: $0.2176 · 1m 37s · 8 turns · 11 in / 5,270 out tokens
With context: $0.3462 · 1m 48s · 13 turns · 17 in / 5,999 out tokens
- Repository
- NeverSight/skills_feed
- Commit
5342bca
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.