On benign tasks containing no malicious skill behavior, SkillSonar preserves high utility on Claude Code (0.817 on GLM-5, 0.836 on Claude Haiku 4.5, and 0.724 on GPT-5.4), demonstrating minimal false-positive disruption of safe execution.

On benign tasks containing no malicious skill behavior, SkillSonar preserves high utility on Claude Code (0.817 on GLM-5, 0.836 on Claude Haiku 4.5, and 0.724 on GPT-5.4), demonstrating minimal false-positive disruption of safe execution. · CiteArk

Measurement results

0/15 assessed

0 supported0 disputed or conflicting15 not assessed

The complete result group stays with its claim. Assessed rows come first; their judgments do not replace the whole-claim conclusion. Expand a row for conditions, sources, and evidence history.

Each row is a measurement with its reported value, latest available observation, and cumulative evidence state.
Metric and conditions	Reported	Latest observation	Evidence judgment
model: GLM-5 · defense: No-guard	0.815 score	—	Not assessed
model: Claude Haiku 4.5 · defense: No-guard	0.827 score	—	Not assessed
model: GPT-5.4 · defense: No-guard	0.793 score	—	Not assessed
model: GLM-5 · defense: System prompt	0.833 score	—	Not assessed
model: Claude Haiku 4.5 · defense: System prompt	0.788 score	—	Not assessed
model: GPT-5.4 · defense: System prompt	0.836 score	—	Not assessed
model: GLM-5 · defense: AcceptEdits	0.796 score	—	Not assessed
model: Claude Haiku 4.5 · defense: AcceptEdits	0.813 score	—	Not assessed
model: GPT-5.4 · defense: AcceptEdits	0.74 score	—	Not assessed
model: GLM-5 · defense: AcceptEdits + allowlist	0.792 score	—	Not assessed
model: Claude Haiku 4.5 · defense: AcceptEdits + allowlist	0.819 score	—	Not assessed
model: GPT-5.4 · defense: AcceptEdits + allowlist	0.764 score	—	Not assessed

On benign tasks containing no malicious skill behavior, SkillSonar preserves high utility on Claude Code (0.817 on GLM-5, 0.836 on Claude Haiku 4.5, and 0.724 on GPT-5.4), demonstrating minimal false-positive disruption of safe execution.

Reported and observed measurements

Assessments (0)

Runs (0)

Measurement results