Compared to action-level runtime defenses AgentSpec and TS-Guard, SkillSonar achieves superior safety–utility operating points on Claude Code with GLM-5: AgentSpec achieves 0.294 ID ASR but drops task utility to 0.581, while TS-Guard achieves 0.412 ID ASR, whereas SkillSonar reaches 0.104 ID ASR with 0.815 task utility.

Compared to action-level runtime defenses AgentSpec and TS-Guard, SkillSonar achieves superior safety–utility operating points on Claude Code with GLM-5: AgentSpec achieves 0.294 ID ASR but drops task utility to 0.581, while TS-Guard achieves 0.412 ID ASR, whereas SkillSonar reaches 0.104 ID ASR with 0.815 task utility. · CiteArk

Measurement results

0/10 assessed

0 supported0 disputed or conflicting10 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
split: id_test · defense: AgentSpec	0.294 fraction	—	Not assessed
split: ood_test · defense: AgentSpec	0.576 fraction	—	Not assessed
split: id_test · defense: AgentSpec	0.581 score	—	Not assessed
defense: AgentSpec	0.642 score	—	Not assessed
split: id_test · defense: TS-Guard	0.412 fraction	—	Not assessed
split: ood_test · defense: TS-Guard	0.492 fraction	—	Not assessed
defense: TS-Guard	0.703 score	—	Not assessed
split: id_test · defense: No-guard	0.477 fraction	—	Not assessed
split: id_test · defense: No-guard	0.829 score	—	Not assessed
defense: No-guard	0.757 score	—	Not assessed

Reported and observed measurements

Assessments (0)

Runs (0)

Measurement results