session-manager

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses local shell commands to gather project context.
  • Evidence: Executes git log --oneline -10 and git status during the session start and sync phases to synchronize the agent's mental model with the repository state.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of ingesting untrusted repository data.
  • Ingestion points: The skill reads project context from CLAUDE.md, files in .claude/plans/, session state in .workflow/active/, and handoff documents in .workflow/handoffs/ (defined in SKILL.md).
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined for the ingested files.
  • Capability inventory: The skill can execute git commands and perform file read/write operations within the workspace.
  • Sanitization: No sanitization or validation of the content within the ingested markdown or JSON files is performed before it is added to the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 05:15 PM
Security Audit — agent-trust-hub — session-manager