Skills
skills/kmshihab7878/claude-code-setup/pruning-view/Gen Agent Trust Hub

pruning-view

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill runs evolve-prune.sh and evolve-regression.sh from ~/.claude/evolution/bin/. These are local scripts within the agent's environment used to generate maintenance reports.
  • [PROMPT_INJECTION]: The skill ingests data from generated reports which may include content from historical records, creating a surface for indirect prompt injection. 1. Ingestion points: reports/*.md files. 2. Boundary markers: Absent. 3. Capability inventory: Reading and summarizing files only; no automated mutations or network exfiltration are authorized by the protocol. 4. Sanitization: None specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 12:08 PM
Security Audit — agent-trust-hub — pruning-view