Audit History
learn-off - 6 audits
Version comparison
Capability and finding changes across audited versions, newest first.
| Version | Date | Result | Review items | Change vs previous |
|---|---|---|---|---|
| v6 Latest | Jul 4, 2026, 01:57 PM | No confirmed findings | 0 | No capability change |
| v5 | Jul 4, 2026, 01:57 PM | No confirmed findings | 0 | External commands Filesystem access |
| v4 | Jun 27, 2026, 01:05 PM | No confirmed findings | 1 | Filesystem access External commands |
| v3 | Jan 16, 2026, 01:01 PM | No confirmed findings | 0 | No capability change |
| v2 | Jan 16, 2026, 01:01 PM | No confirmed findings | 0 | External commands |
| v1 | Jan 10, 2026, 08:57 AM | No confirmed findings | 0 | Baseline |
Jul 4, 2026, 01:57 PM
All seven static findings are false positives caused by Markdown inline code or fenced blocks in SKILL.md. The skill only describes updating a scoped learning-mode state file and does not include executable code, network access, environment access, or prompt-injection text. No additional semantic security findings were found.
Risk Factors
⚙️ External commands (7)
Jul 4, 2026, 01:57 PM
All seven static findings are false positives caused by Markdown inline code or fenced blocks in SKILL.md. The skill only describes updating a scoped learning-mode state file and does not include executable code, network access, environment access, or prompt-injection text. No additional semantic security findings were found.
Risk Factors
⚙️ External commands (7)
Jun 27, 2026, 01:05 PM
Static external command findings are false positives from Markdown inline code, a fenced JSON example, and sample output text in SKILL.md. The weak cryptographic algorithm finding is also a false positive because line 3 is plain frontmatter text. The only confirmed risk is limited filesystem state modification of knowledge/state.json to disable learning mode.
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.
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
📁 Filesystem access (2)
Jan 16, 2026, 01:01 PM
This is a pure prompt-based skill consisting only of documentation. No executable code, no scripts, no network access, no file system access beyond local configuration. The skill provides instructions to modify a local JSON state file and confirm changes to the user. All 16 static findings are false positives caused by keyword pattern matching without semantic context.
Risk Factors
⚙️ External commands (7)
Jan 16, 2026, 01:01 PM
This is a pure prompt-based skill consisting only of documentation. No executable code, no scripts, no network access, no file system access beyond local configuration. The skill provides instructions to modify a local JSON state file and confirm changes to the user. All 16 static findings are false positives caused by keyword pattern matching without semantic context.
Risk Factors
⚙️ External commands (7)
Jan 10, 2026, 08:57 AM
This is a pure prompt-based skill consisting only of documentation. No executable code, no scripts, no network access, no file system access beyond local configuration. The skill provides instructions to modify a local JSON state file and confirm changes to the user.