ns-code-investigator

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of processing untrusted external data.
  • Ingestion points: The skill is instructed to read problems, logs, stack traces, test outputs, and bug descriptions (SKILL.md, Workflow step 1).
  • Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions within the ingested log data.
  • Capability inventory: The agent can execute git status, git diff, and various local test/lint commands (SKILL.md, Workflow step 4).
  • Sanitization: No sanitization or filtering of external content is specified before processing.
  • [COMMAND_EXECUTION]: The skill uses shell commands to investigate the environment and repository state.
  • Evidence: Workflow step 4 explicitly directs the agent to run git status, git diff, and "safe test/lint commands" to gather information.
  • [PROMPT_INJECTION]: The skill contains instructions that reduce user oversight by bypassing permission gates during the investigation phase.
  • Evidence: Workflow step 7 states "Start immediately — no permission gate," which encourages autonomous action without explicit user confirmation for each step.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 07:57 PM
Security Audit — agent-trust-hub — ns-code-investigator