setup-agentic-grimoire
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEPERSISTENCECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PERSISTENCE]: The skill is designed to modify the agent's configuration files,
~/.claude/CLAUDE.mdand~/.codex/AGENTS.md. These files serve as persistent system prompts that influence the agent's behavior across all future work sessions. - [COMMAND_EXECUTION]: The skill provides shell commands and a bash script (
splice.sh) to perform file manipulations. These include creating temporary files, copying existing configurations, and executing a deterministic text splicing logic usingawk. - [PROMPT_INJECTION]: The content defined in the
blocks/directory contains specific instructions that direct the agent's logic, model selection preferences, and mandatory validation steps. By splicing these into the agent's configuration, the skill modifies the agent's core operational instructions.
Audit Metadata