nika-debugging

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the nika command-line utility to perform operations such as listing traces (nika trace ls), replaying journals (nika trace replay), and performing surgical reruns (nika run --from). These commands are used as intended for debugging and forensic purposes.
  • [EXTERNAL_DOWNLOADS]: The skill provides installation instructions via a Homebrew tap (supernovae-st/tap/nika). This resource is owned by the vendor and follows standard software distribution practices.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for processing untrusted data via trace files. Ingestion points: Data is read from .nika/traces/ using forensic tools (SKILL.md). Boundary markers: The system utilizes a permits block to define execution boundaries and nika trace verify to validate a hash-chained journal for tamper evidence. Capability inventory: The toolset includes nika run, nika test, and nika trace evidence for execution and data export. Sanitization: The skill relies on internal hash verification of the trace journal and explicit user-defined permits to mitigate risks from manipulated logs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 09:20 AM
Security Audit — agent-trust-hub — nika-debugging