ss-trace-spec

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute git log --oneline -- openspec/specs to reconstruct file history. This command is restricted to a specific path and is used for read-only metadata retrieval.\n- [PROMPT_INJECTION]: An indirect prompt injection surface exists because the skill reads and summarizes content from external specification files and archives.\n
  • Ingestion points: Data is ingested from the openspec/specs and openspec/changes/archive directories.\n
  • Boundary markers: Absent; the skill does not wrap the ingested file content in delimiters or provide warnings to ignore embedded instructions.\n
  • Capability inventory: The skill's capabilities are limited to reading files and searching git history; it explicitly prohibits file edits and lacks network or administrative tools.\n
  • Sanitization: No validation or sanitization is performed on the content read from the specification files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 03:29 PM
Security Audit — agent-trust-hub — ss-trace-spec