Skills/ClawHub/codex-agent

codex-agent@1.0.0

CAUTION· static only

Publisher claims: "作为项目经理操作 OpenAI Codex CLI 完全体。包含:知识库维护(自动跟踪 Codex 最新功能)、任务执行(提示词设计→执行→监控→质量检查→迭代→汇报)、配置管理(feature flags/模型/skills/MCP)。通过 tmux 操作交互式 TUI,通过 notify hooks + pa..." Observed in sandbox: static analysis only — behavior not yet observed.

Registry ClawHubPublisher henryxiaoyang · publishing for 263 daysInstalls 1,423Last detonated 1 h ago

Coverage

Static only; nothing was run yet.

The label rests on the code, the instruction files and the publisher. The honeypot run replaces this verdict when it completes.

Is [email protected] safe to install?

KillerSkills analysed it statically (a sandbox run is pending) on 2026-11-18 and scored it 67/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Fake instruction boundary tag; Fake instruction boundary tag; Fake instruction boundary tag. Summary: static analysis only — behavior not yet observed.

Behavior trace

Queued for honeypot run · static analysis only
t+CategoryObserved action

This version has not been detonated yet. The score above counts behavior as unknown (50/100), so it cannot read Verified until the honeypot run completes and replaces this verdict.

Static findings

CRITICAL
Fake instruction boundary tagSKILL.md:296 · Fake instruction boundary tagcodex exec --full-auto "<prompt>"Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
CRITICAL
Fake instruction boundary tagSKILL.md:299 · Fake instruction boundary tagcodex exec --full-auto -i image.png "<prompt>"Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
CRITICAL
Fake instruction boundary tagSKILL.md:297 · Fake instruction boundary tagcodex exec --full-auto -m gpt-5.2 -C /path "<prompt>"Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
CRITICAL
Fake instruction boundary tagSKILL.md:298 · Fake instruction boundary tagcodex exec --full-auto -c 'model_reasoning_effort="xhigh"' "<prompt>"Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
CRITICAL
Fake instruction boundary tagSKILL.md:87 · Fake instruction boundary tagnohup codex exec --full-auto -C <workdir> "<prompt>" > /tmp/codex_exec_output.txt 2>&1 &Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Python subprocess execution detectedhooks/on_complete.py:65 · Python subprocess execution detectedproc = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedhooks/on_complete.py:37 · Python subprocess execution detectedproc = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Process kill commandhooks/stop_codex.sh:12 · Process kill command — agent can terminate processeskill "$(cat "$MONITOR_PID_FILE")" 2>/dev/null || trueFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05

Cost to the agent

Instructions loaded
≈1,752 tok
1 file · 21 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.

What would make this Verified

13 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim henryxiaoyang: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
  3. Process kill command≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
Publish a new version to ClawHub, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim henryxiaoyangPropose a patch to the maintainer