Audit History
azure-cost - 5 audits
Version comparison
Capability and finding changes across audited versions, newest first.
| Version | Date | Result | Review items | Change vs previous |
|---|---|---|---|---|
| v5 Latest | Jul 23, 2026, 06:20 PM | 2 confirmed | 1 | No capability change |
| v4 | Jul 8, 2026, 08:46 AM | 1 confirmed | 0 | No capability change |
| v3 | Jul 5, 2026, 09:04 PM | No confirmed findings | 0 | No capability change |
| v2 | Jun 30, 2026, 08:10 AM | 1 confirmed | 4 | Filesystem accessEnv variables |
| v1 | Apr 9, 2026, 08:42 AM | No confirmed findings | 2 | Baseline |
Jul 23, 2026, 06:20 PM
Review confirmed one high-risk cleanup command that recursively removes a relative temp directory without proving the skill owns it. Two semantic issues remain: unqualified resource deletion advice and budget creation without an explicit approval checkpoint. The other 79 static findings are documentation false positives involving Markdown links, code fences, official URLs, and non-secret environment selectors.
Confirmed security concerns (2)
Capability review items (1)
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
📁 Filesystem access (12)
⚙️ External commands (24)
🔑 Env variables (13)
Jul 8, 2026, 08:46 AM
The static findings are mostly documentation false positives: Markdown links, Azure API examples, official Azure URLs, and credential best-practice snippets. No prompt injection or credential exfiltration evidence was found. One semantic issue remains because AKS guidance includes a state-changing budget creation command that should require explicit confirmation.
Confirmed security concerns (1)
Risk Factors
📁 Filesystem access (12)
⚙️ External commands (24)
🔑 Env variables (13)
Jul 5, 2026, 09:04 PM
No prompt-injection language, malicious data exfiltration, or business-logic abuse was found in the reviewed files. The static findings are false positives from Markdown links, code fences, official Azure endpoints, Azure CLI examples, and credential best-practice snippets.
Risk Factors
📁 Filesystem access (12)
⚙️ External commands (35)
🔑 Env variables (13)
🌐 Network access (2)
Jun 30, 2026, 08:10 AM
Static analysis flagged many command, network, filesystem, environment, and weak-crypto patterns, but most are false positives caused by markdown examples, Azure API field names, and relative documentation links. No prompt injection or confirmed malicious exfiltration was found. The remaining risk is moderate because the skill can guide authenticated Azure CLI, AKS, and Redis operations, including cost-changing updates and deletion recommendations that require explicit user approval.
Confirmed security concerns (1)
Capability review items (4)
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 (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 (4)
🌐 Network access (3)
📁 Filesystem access (3)
Apr 9, 2026, 08:42 AM
Static analysis flagged 888 pattern matches (risk score 100/100), but evaluation confirms these are FALSE POSITIVES. The skill is documentation-based markdown files containing Azure Cost Management API usage examples. Backtick patterns are PowerShell command formatting, not shell execution. Path sequences are relative documentation links. URLs are legitimate Azure references. No malicious intent detected. Minor risk from documented CLI command examples that require Azure authentication.
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.