Skills
Navigate
skills/full-stack-skills/speckit-skills/openspec-archive-change/Gen Agent Trust Hub

openspec-archive-change

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to run openspec CLI commands and standard filesystem utilities like mkdir and mv to manage and archive project changes.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and obey instructions from external data sources, which presents a surface for potential indirect prompt injection.\n
  • Ingestion points: The skill reads JSON data from openspec instructions archive and openspec instructions specs, and parses the content of tasks.md.\n
  • Boundary markers: The instructions explicitly direct the agent to treat ingested context as a required prompt-level input and to apply rules from the data to its file-writing operations.\n
  • Capability inventory: The agent can execute shell commands via the openspec tool and modify project specification files.\n
  • Sanitization: There is no structured validation or filtering of the ingested instructions beyond a general guidance to report conflicts and avoid verbatim copying of data into outputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 06:49 AM