Enforces planning-only roles and validates prompt files for Finesse.
npx skills add https://github.com/jonathanung/finesse --skill planner-identity
Please help me install this Agent Skill. Skill: planner-identity Source: https://github.com/jonathanung/finesse/tree/main/plugins/finesse/skills/planner-identity Command: npx skills add https://github.com/jonathanung/finesse --skill planner-identity
This Skill enforces non-negotiable rules and identity constraints for all Finesse planning commands, ensuring consistent and safe AI behavior during task planning.
Use the planner-identity skill to ensure all Finesse planning adheres to strict safety and role guidelines.
High-intent search queries and answers about installing and using this skill.
To enforce a planning-only role for an AI agent, apply strict identity constraints and safety guardrails. This ensures the agent focuses solely on generating validated prompt files and never attempts unauthorized code execution or data exfiltration.
AI agent safety guardrails for task planning work by enforcing a strict planning-only identity, mandating codebase exploration before solution design, and validating all outputs as prompt files to prevent unauthorized modifications.
Mandate codebase exploration by configuring the agent's core identity to enforce strict adherence to predefined planning phases. This guarantees the AI analyzes the existing codebase before designing any solutions.
Yes, you can restrict an AI agent to generate planning prompts without allowing direct code changes by enforcing a planning-only identity. This prevents unauthorized code execution and ensures all outputs are validated prompt files.
An AI agent skips planning phases or executes code prematurely when it lacks strict identity constraints and operational rules. Enforcing predefined planning phases and a planning-only role ensures consistent, safe behavior during task planning.
The main limitation of a planning-only AI agent is that it cannot execute code or make direct changes to the codebase. Its strict operational rules restrict all outputs to validated prompt files, preventing any actual implementation.
vasilyu1983Automate design and deployment of production-grade AI agents.
vultukGenerate Ralph Wiggum prompts for Claude Code iterative loops.
T1nker-1220Restrict AI actions to read-only operations and real data during development workflows.
thangvawnIntegrate AI agents into engineering workflows for code generation and review.
