paw-ps-software-executor
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill demonstrates legitimate functionality focused on product design and documentation. It contains no evidence of malicious intent, obfuscation, or persistence mechanisms.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests data from local curated memory files such as curated/product-context.md (Ingestion points: SKILL.md). There are no explicit boundary markers or sanitization procedures described. However, because the skill's capabilities are strictly limited to writing documentation artifacts (Capability inventory: references/save-memory.md), this ingestion surface does not present a significant security risk.
- [DATA_EXFILTRATION]: The skill operates entirely within the designated .pawbytes project workspace and does not perform network operations or access sensitive system-level credentials or files.
Audit Metadata