skills/affaan-m/ecc/unified-memory/Gen Agent Trust Hub

unified-memory

Pass

Audited by Gen Agent Trust Hub on Jul 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the ecc-universal package from the npm registry to provide the core CLI and MCP functionality.
  • [COMMAND_EXECUTION]: The skill executes the ecc memory CLI and ecc-memory-mcp server to manage, search, and retrieve context documents from the local filesystem.
  • [PROMPT_INJECTION]: The skill identifies a potential surface for indirect prompt injection when ingesting shared memory documents.
  • Ingestion points: The memory_read and memory_search tools ingest Markdown documents from the local filesystem into the agent's context (SKILL.md).
  • Boundary markers: Present; the instructions explicitly direct the agent to treat recalled memory bodies as untrusted context and never as executable instructions (SKILL.md).
  • Capability inventory: The skill provides capabilities to read, search, and write Markdown files to local repository paths or the user home directory (SKILL.md).
  • Sanitization: Present; the documentation notes that the runtime includes a backstop filter to reject known secret shapes during memory storage (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 26, 2026, 11:14 AM
Security Audit — agent-trust-hub — unified-memory