fraim-framework 2.0.36 → 2.0.37

package/registry/agent-guardrails.md

@@ -1,63 +1,63 @@
-# AI Agent Guardrails
-
-This file references the centralized rules located in `rules/` to ensure consistency across all AI platforms.
-
-## SUCCESS CRITERIA (THE "FRAIM 5")
-**Source**: Retrieve via `get_fraim_file({ path: "rules/agent-success-criteria.md" })`
-
-All agents are evaluated on:
-1. **Integrity (100 pts)**: Honesty above all.
-2. **Correctness (50 pts)**: Architecture & tests.
-3. **Completeness (30 pts)**: Thoroughness.
-4. **Independence (20 pts)**: Smart decisions.
-5. **Speed (10 pts)**: Velocity.
-
-## Referenced Rules
-
-### 0. Integrity
-**Source**: Retrieve via `get_fraim_file({ path: "rules/integrity-and-test-ethics.md" })`
-
-THIS IS THE MOST CRITICAL RULE. Be ethical, truthful, honest above all.
-
-### 1. Simplicity
-**Source**: Retrieve via `get_fraim_file({ path: "rules/simplicity.md" })`
-
-Keep solutions simple and focused, avoid over-engineering. Focus on the assigned issue only and don't make unrelated changes.
-
-### 2. Communication  
-**Source**: Retrieve via `get_fraim_file({ path: "rules/communication.md" })`
-
-Establish clear communication patterns and progress reporting standards for effective coordination between agents and stakeholders.
-
-### 3. Architecture
-**Source**: Retrieve via `get_fraim_file({ path: "rules/architecture.md" })`
-
-Maintain clean architectural boundaries by using BAML (LLM) for natural-language understanding and TypeScript for deterministic work.
-
-### 4. Continuous Learning
-**Source**: Retrieve via `get_fraim_file({ path: "rules/continuous-learning.md" })`
-
-Prevent repeating past mistakes by systematically learning from retrospectives, RFCs, and historical issue patterns.
-
-### 5. Agent Testing Guidelines
-**Source**: Retrieve via `get_fraim_file({ path: "rules/agent-testing-guidelines.md" })`
-
-Comprehensive testing and validation requirements with concrete evidence. Ensures all work is thoroughly validated before completion.
-
-### 6. Local Development
-**Source**: Retrieve via `get_fraim_file({ path: "rules/local-development.md" })`
-
-Local development guidelines and workspace safety. Enables safe parallel development through strict workspace separation.
-
-### 7. Software Development Lifecycle
-**Source**: Retrieve via `get_fraim_file({ path: "rules/software-development-lifecycle.md" })`
-
-### 9. Merge Requirements
-**Source**: Retrieve via `get_fraim_file({ path: "rules/merge-requirements.md" })`
-
-Enforces a strict `git rebase` workflow to ensure feature branches are up-to-date with `master` before merging, maintaining a clean and stable history.
-
-### 10. Best practices while debuggin
-**Source**: Retrieve via `get_fraim_file({ path: "rules/successful-debugging-patterns.md" })`
-
+# AI Agent Guardrails
+
+This file references the centralized rules located in `rules/` to ensure consistency across all AI platforms.
+
+## SUCCESS CRITERIA (THE "FRAIM 5")
+**Source**: Retrieve via `get_fraim_file({ path: "rules/agent-success-criteria.md" })`
+
+All agents are evaluated on:
+1. **Integrity (100 pts)**: Honesty above all.
+2. **Correctness (50 pts)**: Architecture & tests.
+3. **Completeness (30 pts)**: Thoroughness.
+4. **Independence (20 pts)**: Smart decisions.
+5. **Speed (10 pts)**: Velocity.
+
+## Referenced Rules
+
+### 0. Integrity
+**Source**: Retrieve via `get_fraim_file({ path: "rules/integrity-and-test-ethics.md" })`
+
+THIS IS THE MOST CRITICAL RULE. Be ethical, truthful, honest above all.
+
+### 1. Simplicity
+**Source**: Retrieve via `get_fraim_file({ path: "rules/simplicity.md" })`
+
+Keep solutions simple and focused, avoid over-engineering. Focus on the assigned issue only and don't make unrelated changes.
+
+### 2. Communication  
+**Source**: Retrieve via `get_fraim_file({ path: "rules/communication.md" })`
+
+Establish clear communication patterns and progress reporting standards for effective coordination between agents and stakeholders.
+
+### 3. Architecture
+**Source**: Retrieve via `get_fraim_file({ path: "rules/architecture.md" })`
+
+Maintain clean architectural boundaries by using BAML (LLM) for natural-language understanding and TypeScript for deterministic work.
+
+### 4. Continuous Learning
+**Source**: Retrieve via `get_fraim_file({ path: "rules/continuous-learning.md" })`
+
+Prevent repeating past mistakes by systematically learning from retrospectives, RFCs, and historical issue patterns.
+
+### 5. Agent Testing Guidelines
+**Source**: Retrieve via `get_fraim_file({ path: "rules/agent-testing-guidelines.md" })`
+
+Comprehensive testing and validation requirements with concrete evidence. Ensures all work is thoroughly validated before completion.
+
+### 6. Local Development
+**Source**: Retrieve via `get_fraim_file({ path: "rules/local-development.md" })`
+
+Local development guidelines and workspace safety. Enables safe parallel development through strict workspace separation.
+
+### 7. Software Development Lifecycle
+**Source**: Retrieve via `get_fraim_file({ path: "rules/software-development-lifecycle.md" })`
+
+### 9. Merge Requirements
+**Source**: Retrieve via `get_fraim_file({ path: "rules/merge-requirements.md" })`
+
+Enforces a strict `git rebase` workflow to ensure feature branches are up-to-date with `master` before merging, maintaining a clean and stable history.
+
+### 10. Best practices while debuggin
+**Source**: Retrieve via `get_fraim_file({ path: "rules/successful-debugging-patterns.md" })`
+
 Patterns on debugging issues systematically and converting learnings into test cases