Audit History
audit - 10 audits
Version comparison
Capability and finding changes across audited versions, newest first.
| Version | Date | Result | Review items | Change vs previous |
|---|---|---|---|---|
| v10 Latest | 2026年7月9日 下午02:37 | No confirmed findings | 2 | No capability change |
| v9 | 2026年7月9日 下午02:37 | No confirmed findings | 2 | No capability change |
| v8 | 2026年7月5日 上午09:07 | No confirmed findings | 2 | No capability change |
| v7 | 2026年7月5日 上午09:07 | No confirmed findings | 2 | No capability change |
| v6 | 2026年6月29日 上午09:14 | 1 confirmed | 0 | No capability change |
| v5 | 2026年1月16日 下午10:52 | No confirmed findings | 0 | No capability change |
| v4 | 2026年1月16日 下午10:52 | No confirmed findings | 0 | External commands |
| v3 | 2026年1月10日 下午01:05 | No confirmed findings | 0 | No capability change |
| v2 | 2026年1月10日 下午01:05 | No confirmed findings | 0 | No capability change |
| v1 | 2026年1月10日 下午01:05 | No confirmed findings | 0 | Baseline |
2026年7月9日 下午02:37
No prompt injection, data exfiltration, or malicious intent was found in SKILL.md. Three static detections are markdown-only false positives. Two detections are real external-command guidance for local validation tools.
Capability review items (2)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (5)
2026年7月9日 下午02:37
No prompt injection, data exfiltration, or malicious intent was found in SKILL.md. Three static detections are markdown-only false positives. Two detections are real external-command guidance for local validation tools.
Capability review items (2)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (5)
2026年7月5日 上午09:07
Most static findings are Markdown formatting, not Ruby or shell backtick execution. Two findings are confirmed because the skill asks the agent to run local project commands such as tsc and npm lint scripts. No prompt injection or data exfiltration intent was found in SKILL.md.
Capability review items (2)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (5)
2026年7月5日 上午09:07
Most static findings are Markdown formatting, not Ruby or shell backtick execution. Two findings are confirmed because the skill asks the agent to run local project commands such as tsc and npm lint scripts. No prompt injection or data exfiltration intent was found in SKILL.md.
Capability review items (2)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (5)
2026年6月29日 上午09:14
Static analysis flagged external command patterns and weak cryptography patterns in SKILL.md. Manual review confirmed only local command execution guidance for type checking and linting; the weak cryptography matches are false positives in prose and markdown examples.
Confirmed security concerns (1)
Static false positives ignored (2)
These static matches were dismissed by semantic review or matched schema-only tokens, so they are shown for transparency but do not drive the quality score.
Risk Factors
⚙️ External commands (2)
Detected Patterns
2026年1月16日 下午10:52
Pure prompt-based skill containing only markdown instructions for systematic code audit methodology. Contains no executable code, network calls, or file system modifications. The static analyzer flagged benign content (documentation, command references, and metadata) as false positives. All 17 detected patterns are harmless - this is a read-only analysis tool.
Risk Factors
⚙️ External commands (6)
2026年1月16日 下午10:52
Pure prompt-based skill containing only markdown instructions for systematic code audit methodology. Contains no executable code, network calls, or file system modifications. The static analyzer flagged benign content (documentation, command references, and metadata) as false positives. All 17 detected patterns are harmless - this is a read-only analysis tool.
Risk Factors
⚙️ External commands (6)
2026年1月10日 下午01:05
Pure prompt-based skill providing code audit methodology guidelines. Contains no executable code, only markdown instructions for systematic file review and analysis. Uses standard development tools (tsc, eslint, grep) for read-only analysis.
2026年1月10日 下午01:05
Pure prompt-based skill providing code audit methodology guidelines. Contains no executable code, only markdown instructions for systematic file review and analysis. Uses standard development tools (tsc, eslint, grep) for read-only analysis.
2026年1月10日 下午01:05
Pure prompt-based skill providing code audit methodology guidelines. Contains no executable code, only markdown instructions for systematic file review and analysis. Uses standard development tools (tsc, eslint, grep) for read-only analysis.