Skills
skills/jankneumann/agentic-coding-tools/linear-explore-feature/Gen Agent Trust Hub

linear-explore-feature

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes the make architecture command to generate architecture documentation if specific analysis files are missing from the project directory. \n- [COMMAND_EXECUTION]: Utilizes the openspec CLI tool and a local Python script scripts/coordination_bridge.py for state gathering and historical data recall. \n- [COMMAND_EXECUTION]: Performs dynamic execution by identifying and loading additional skills from computed filesystem paths such as .claude/commands/opsx/ and .gemini/skills/. \n- [PROMPT_INJECTION]: Susceptible to indirect prompt injection via the ingestion of untrusted data from codebase specifications and architecture files. \n
  • Ingestion points: Output from openspec list and JSON files in docs/architecture-analysis/. \n
  • Boundary markers: None identified. \n
  • Capability inventory: Local subprocess execution, dynamic skill loading, and file-write operations. \n
  • Sanitization: No explicit validation or filtering of ingested data before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 04:38 AM