Audit History
gws-slides - 7 audits
Version comparison
Capability and finding changes across audited versions, newest first.
| Version | Date | Result | Review items | Change vs previous |
|---|---|---|---|---|
| v7 Latest | Aug 5, 2026, 11:31 AM | No confirmed findings | 0 | No capability change |
| v6 | Jul 9, 2026, 11:35 AM | No confirmed findings | 0 | No capability change |
| v5 | Jul 9, 2026, 11:35 AM | No confirmed findings | 0 | No capability change |
| v4 | Jul 5, 2026, 03:45 PM | No confirmed findings | 2 | No capability change |
| v3 | Jul 5, 2026, 03:45 PM | No confirmed findings | 2 | No capability change |
| v2 | Jun 30, 2026, 02:04 AM | No confirmed findings | 3 | External commandsNetwork accessFilesystem access |
| v1 | Apr 13, 2026, 08:52 AM | No confirmed findings | 0 | Baseline |
Aug 5, 2026, 11:31 AM
All 12 static findings are false positives caused by Markdown backticks, prose containing request-related words, and a fixed sibling documentation path. No malicious intent, prompt injection, unsafe dynamic command construction, or data exfiltration behavior was found.
Risk Factors
⚙️ External commands (10)
🌐 Network access (1)
📁 Filesystem access (1)
Jul 9, 2026, 11:35 AM
Static findings were reviewed against SKILL.md and appear to be Markdown command examples, inline command names, and one literal shared-skill reference. I found no evidence of prompt injection, data exfiltration intent, credential theft, or executable code embedded in the skill file.
Risk Factors
⚙️ External commands (10)
🌐 Network access (1)
📁 Filesystem access (1)
Jul 9, 2026, 11:35 AM
Static findings were reviewed against SKILL.md and appear to be Markdown command examples, inline command names, and one literal shared-skill reference. I found no evidence of prompt injection, data exfiltration intent, credential theft, or executable code embedded in the skill file.
Risk Factors
⚙️ External commands (10)
🌐 Network access (1)
📁 Filesystem access (1)
Jul 5, 2026, 03:45 PM
The markdown backtick, command example, and API description detections are mostly false positives. One high-severity filesystem issue remains because SKILL.md line 16 instructs agents to read ../gws-shared/SKILL.md outside the reviewed skill package. No prompt injection or data exfiltration intent was found in the reviewed file.
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 (10)
🌐 Network access (1)
📁 Filesystem access (1)
Jul 5, 2026, 03:45 PM
The markdown backtick, command example, and API description detections are mostly false positives. One high-severity filesystem issue remains because SKILL.md line 16 instructs agents to read ../gws-shared/SKILL.md outside the reviewed skill package. No prompt injection or data exfiltration intent was found in the reviewed file.
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 (10)
🌐 Network access (1)
📁 Filesystem access (1)
Jun 30, 2026, 02:04 AM
Static analysis reported command execution, network, filesystem, and weak cryptography patterns in SKILL.md. Review found no prompt injection, malware, or cryptographic code; most high-severity hits are false positives from Markdown text. Residual risk remains because the skill instructs use of the authenticated gws CLI, which can read, create, and update Google Slides data.
Capability review items (3)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Static false positives ignored (1)
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 (4)
🌐 Network access (1)
📁 Filesystem access (1)
Detected Patterns
Apr 13, 2026, 08:52 AM
All 23 static analysis findings are false positives. The scanner incorrectly treated markdown documentation as executable code. The SKILL.md file contains only usage instructions and CLI command examples for the Google Workspace Slides tool. There is no executable code, no network calls, no filesystem operations, and no security-sensitive patterns. The skill is a legitimate productivity tool from the Google Workspace CLI repository.
Static false positives ignored (4)
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.