Snapshots local Codex/Claude CLI sessions across terminals and restores IDs in Ghostty after reboot (保存会话/重启前备份/恢复窗口), audits survivors, and makes change-only automatic backups (定期增量快照). Handles unmanaged Codex account switches (额度耗尽换号); Minions native_runtime uses Account Manager. Not --resume help, screenshots or tmux.
- Stars
- 1,534
- Version
- v2
- Copies
- 0
Overview
Waxell rates it high risk: 48 findings, highest severity high.
- Highest severity
- high
- Findings
- 48
- What ran
- Static analysis + deep scan
A clean result means nothing alarming in what is published, not a guarantee. How the checks work
We don't republish the skill's instructions. Read them on GitHub at the reviewed commit.
Security
Security scan
Checked yesterday
Result
High risk Waxell rates it high risk: 48 findings, highest severity high.
Checked with static analysis + deep scan: servers it launches, packages it installs at runtime, scripts piped to a shell and bundled binaries. The deep scan also read the instructions for prompt injection and data exfiltration. Nothing was run. How the checks work
Findings (48)
high · 6
- high
analysis-evasion: Incomplete referenced artifact analysis
SKILL.md:129 · skillspector:AE1
- high
Data Exfiltration: Env Variable Harvesting
scripts/ghostty_session.py:902 · skillspector:E2
- high
Privilege Escalation: Credential Access
scripts/ghostty_switch.py:43 · skillspector:PE3
- high
Agent Snooping: Agent Config Directory Access
scripts/test_ghostty_session.py:270 · skillspector:AS1
- high
Agent Snooping: Agent Config Directory Access
scripts/test_ghostty_session.py:273 · skillspector:AS1
- high
Agent Snooping: Agent Config Directory Access
scripts/test_ghostty_session.py:283 · skillspector:AS1
medium · 42
- medium
Without declared permissions the skill's intent is opaque and cannot be validated.
SKILL.md:1 · skillspector:LP3
- medium
Rogue Agent: Session Persistence
SKILL.md:177 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
references/automatic-snapshots.md:51 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
references/automatic-snapshots.md:67 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:14 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:658 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:668 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:683 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:697 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:710 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:711 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:731 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/test_ghostty_watch.py:11 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/test_ghostty_watch.py:380 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/test_ghostty_watch.py:788 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/test_ghostty_watch.py:865 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/test_ghostty_watch.py:875 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
scripts/test_ghostty_watch.py:895 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
references/codex-account-switch.md:83 · skillspector:RA2
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_native_host.py:30 · skillspector:AST4
- medium
Data Flow: Tainted flow: 'argv' from pathlib.Path.read_bytes (line 170, file read) → subprocess.run (code execution)
scripts/ghostty_native_host.py:30 · skillspector:TT2
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_native_host.py:66 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_native_host.py:141 · skillspector:AST4
- medium
Data Flow: Tainted flow: 'terminal_id' from pathlib.Path.read_bytes (line 215, file read) → subprocess.run (code executi
scripts/ghostty_native_host.py:141 · skillspector:TT2
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_native_host.py:160 · skillspector:AST4
- medium
Data Flow: Tainted flow: 'marker' from pathlib.Path.read_bytes (line 212, file read) → subprocess.run (code execution)
scripts/ghostty_native_host.py:160 · skillspector:TT2
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_native_host.py:374 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_session.py:194 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_session.py:215 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_session.py:783 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_session.py:1019 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_session_port.py:102 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_switch.py:333 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_switch.py:548 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_watch.py:57 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_watch.py:672 · skillspector:AST4
- medium
Rogue Agent: Session Persistence
scripts/ghostty_watch.py:735 · skillspector:RA2
- medium
Dangerous Code Execution: subprocess module call
scripts/ghostty_watch.py:818 · skillspector:AST4
- medium
Dangerous Code Execution: Dynamic import via __import__()
scripts/test_ghostty_session.py:424 · skillspector:AST3
- medium
Data Flow: Tainted flow: 'record' from pathlib.Path.read_text (line 295, file read) → pathlib.Path.write_text (file writ
scripts/test_ghostty_switch.py:297 · skillspector:TT2
- medium
Dangerous Code Execution: subprocess module call
scripts/test_ghostty_switch.py:764 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/test_ghostty_switch.py:777 · skillspector:AST4
Versions
Version history
Each distinct content we have seen, newest first, with the verdict it got at the time.
- v22026-11-166 added, 11 changed High riskgot riskier
- v12026-11-03 No issues
Files
Files
18 files, 463 KB in total. Names only; open them on GitHub.
- SKILL.md
- references/account-manager-terminal-port.md
- references/automatic-snapshots.md
- references/codex-account-switch.md
- references/restore_mechanics.md
- references/session_liveness.md
- scripts/ghostty_ax_read.swift
- scripts/ghostty_native_host.py
- scripts/ghostty_session.py
- scripts/ghostty_session_port.py
- scripts/ghostty_storage.py
- scripts/ghostty_switch.py
- scripts/ghostty_watch.py
- scripts/test_ghostty_native_host.py
- scripts/test_ghostty_session.py
- scripts/test_ghostty_session_port.py
- scripts/test_ghostty_switch.py
- scripts/test_ghostty_watch.py
See and control which skills your people use
Inside your organization, Waxell scans each skill and plugin before anyone installs it, approves what passes and publishes it to a private marketplace, and puts every MCP tool call behind one governed URL.
30 minutes. No commitment.
SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF
More engineering skills and plugins
git-safety-netdaymadeAudits, preserves and recovers local Git state before cleanup: unpushed or wrong-branch commits, dirty worktrees, duplicate clones,… High riskEngineeringAI & agents1.5k
github-opsdaymadeOperates GitHub via gh CLI and REST/GraphQL: PRs, issues, Actions, self-hosted Runner inventory/capacity, repos, collaborators, org… High riskEngineeringDesign1.5k
lark-cli-routerdaymadeRoutes any Feishu/Lark/Doubao operation, including recording lookup by date (按日期找会议录音), through the version-matched lark-cli domain guide… High riskEngineeringProductivity1.5k
openclaw-model-switchdaymadeSwitches or repairs an OpenClaw instance's model config: changes the default model, adds model definitions, and fixes 401 "Invalid token",… High riskEngineeringAI & agents1.5k