The Agent Skills Directory

[SAFE]: The skill implements auditing logic using standard, non-destructive search tools such as find, grep, rg (ripgrep), and awk to scan the local filesystem for shell script patterns.
[SAFE]: Dependencies are restricted to internal platform modules (imbue:proof-of-work, imbue:review-core, imbue:structured-output) and local markdown modules.
[SAFE]: The verification process utilizes a local Python script (plugins/imbue/scripts/citation_verifier.py) to confirm that audit findings match actual code locations, which serves as a safeguard against inaccurate reporting.
[SAFE]: No network communication, unauthorized data access, obfuscation, or privilege escalation mechanisms were detected in the skill instructions or modules.

shell-review