user-onboarding

Pass

Audited by ZeroLeaks on Apr 15, 2026

Risk Level: NONE
Scan Summary

The SKILL.md is clear, fully declarative, and easy to review—it describes a conversational advisory skill about product onboarding design with no tool invocations, file operations, or external fetches. It stays clear of instruction/data boundary issues and does not push the agent to treat user-supplied content as trusted policy. The tested scenarios did not show material prompt-injection risk or skill-induced worsening of downstream behavior. Confidence is low not due to the skill's content—which looks clean—but because behavior analysis was not run, leaving a gap in validating whether the skill materially changes downstream agent behavior compared to baseline.

Score
93/100
Verdict
PASS
Confidence
low
Findings
0
Section Analysis (3)
TransparencyPASS
96/100

This is a purely advisory/conversational skill about product onboarding design. It contains no tool invocations, no shell commands, no remote fetches, no file writes, no external communication, and no hidden execution paths. All behavior is fully described in the markdown.

Prompt InjectionPASS
92/100

The scanned skill keeps data and instructions reasonably separate and does not strongly encourage the agent to treat external content as policy.

Agent BehaviorSkipped

Behavior analysis was not run.

Checks
TransparencyNo hidden execution paths or embedded secrets
Prompt injectionData and instructions kept separate
Agent behaviorNo harmful downstream agent effects detected
Coverage DetailsClick to expand
Discovered Files
1
Omitted Files
0
Analyzed Bytes
3.4 KB
Sections Completed
2
Sections Skipped
1
Behavior Probes
0/0
Audit Metadata
Score
93/100
Verdict
PASS
Confidence
low
Sections
2/3 completed
Mode
risk
Files Scanned
1
Duration
66.4s
Analyzed
Apr 15, 2026, 11:09 PM
Security Audit — zeroleaks — user-onboarding