Skills
Navigate
skills/bitwarden/ai-plugins/extracting-session-data/Gen Agent Trust Hub

extracting-session-data

Warn

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script filter-sessions.sh is vulnerable to shell command injection in the date_to_timestamp function, which processes the --since and --until arguments. A maliciously crafted date string can break out of the command substitution and execute arbitrary shell commands with the user's permissions.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes session logs that may contain untrusted data from previous agent interactions.
  • Ingestion points: Session log files in ~/.claude/projects/ are processed by extract-data.sh, filter-sessions.sh, and list-sessions.sh.
  • Boundary markers: Absent. Data is returned as raw text without delimiters.
  • Capability inventory: Read access to session logs; processing via shell utilities including jq, grep, awk, sed, date, bc, stat, and du across all scripts.
  • Sanitization: The skill uses jq for structural extraction but does not sanitize or escape the content of log messages.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 17, 2026, 03:53 PM