skills/zzci/skills/pma-mem/Gen Agent Trust Hub

pma-mem

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for using shell commands including curl, jq, gh, and md5sum to interact with external APIs and perform data processing tasks such as deduplication and formatting.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with external services (Memos, BKD, GitHub, Linear) via REST APIs to fetch and store knowledge points. These operations are performed using user-provided environment variables for endpoints and authentication tokens.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it is designed to ingest and process untrusted text (comments, logs, issue bodies) from external task tracking systems during classification.
  • Ingestion points: Issue logs and conversation history retrieved from BKD, GitHub, and Linear APIs as described in references/knowledge-sync.md.
  • Boundary markers: The instructions do not specify the use of explicit delimiters or "ignore embedded instructions" warnings when processing the extracted content.
  • Capability inventory: The agent has the ability to execute shell commands, perform network requests via curl, and interact with repositories using the GitHub CLI.
  • Sanitization: There is no evidence of explicit sanitization or filtering of external content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 09:00 AM
Security Audit — agent-trust-hub — pma-mem