Audit History
claude-opus-4-5-migration - 10 audits
Version comparison
Capability and finding changes across audited versions, newest first.
| Version | Date | Result | Review items | Change vs previous |
|---|---|---|---|---|
| v10 Latest | Jul 23, 2026, 12:26 PM | No confirmed findings | 0 | No capability change |
| v9 | Jul 8, 2026, 12:40 AM | No confirmed findings | 0 | No capability change |
| v8 | Jul 6, 2026, 06:14 AM | No confirmed findings | 0 | External commands |
| v7 | Jun 28, 2026, 10:07 PM | No confirmed findings | 0 | No capability change |
| v6 | Jan 21, 2026, 02:51 PM | No confirmed findings | 0 | External commands |
| v5 | Jan 16, 2026, 08:37 PM | No confirmed findings | 0 | No capability change |
| v4 | Jan 16, 2026, 08:37 PM | No confirmed findings | 0 | External commands |
| v3 | Jan 9, 2026, 02:54 PM | No confirmed findings | 0 | No capability change |
| v2 | Jan 9, 2026, 02:54 PM | No confirmed findings | 0 | No capability change |
| v1 | Jan 9, 2026, 02:54 PM | No confirmed findings | 0 | Baseline |
Jul 23, 2026, 12:26 PM
All 30 static findings are false positives. The six blocker matches are ordinary coding and design guidance, while the 24 command matches are Markdown delimiters, identifiers, or reference links with no execution behavior. No prompt injection or other semantic security risk was found.
Risk Factors
⚙️ External commands (24)
Jul 8, 2026, 12:40 AM
All 30 static findings were adjudicated as false positives. The SKILL.md matches are Markdown formatting and example model strings, while the prompt-snippet matches are migration guidance rather than reconnaissance or prompt injection.
Risk Factors
⚙️ External commands (24)
Jul 6, 2026, 06:14 AM
All static findings were adjudicated as false positives. The external command hits are markdown backticks, code fences, model identifiers, and prompt examples, not executable shell calls. The reconnaissance hits are prompt migration snippets with no evidence of prompt injection, data exfiltration, or host inspection intent.
Risk Factors
⚙️ External commands (24)
Jun 28, 2026, 10:07 PM
Official Anthropic skill with documentation-only migration guidance and no executable scripts. Static external-command indicators are false positives from code examples and prompt text, not operational command execution.
Jan 21, 2026, 02:51 PM
Official Anthropic skill for safe migration between Claude models. Performs only text-based replacements in user code. No dynamic execution, no external dependencies, no network calls. Risk factors detected are false positives from static analysis of markdown documentation.
Jan 16, 2026, 08:37 PM
AI analysis failed after multiple attempts - MANUAL REVIEW REQUIRED before publishing. This skill cannot be auto-published until reviewed by a human.
Risk Factors
⚙️ External commands (75)
Detected Patterns
Jan 16, 2026, 08:37 PM
AI analysis failed after multiple attempts - MANUAL REVIEW REQUIRED before publishing. This skill cannot be auto-published until reviewed by a human.
Risk Factors
⚙️ External commands (75)
Detected Patterns
Jan 9, 2026, 02:54 PM
Official Anthropic skill containing only documentation and migration guidance. No executable code, network operations, or filesystem modifications. Safe for publishing.
Jan 9, 2026, 02:54 PM
Official Anthropic skill containing only documentation and migration guidance. No executable code, network operations, or filesystem modifications. Safe for publishing.
Jan 9, 2026, 02:54 PM
Official Anthropic skill containing only documentation and migration guidance. No executable code, network operations, or filesystem modifications. Safe for publishing.