-
Notifications
You must be signed in to change notification settings - Fork 0
Implement personality injection and system prompt construction #13
Copy link
Copy link
Closed
Labels
prio:highImportant, should be prioritizedImportant, should be prioritizedscope:medium1-3 days of work1-3 days of workspec:agent-systemDESIGN_SPEC Section 3 - Agent SystemDESIGN_SPEC Section 3 - Agent Systemtype:featureNew feature implementationNew feature implementation
Milestone
Description
Context
Translate agent personality traits and configuration into effective system prompts per spec 3.1. The prompt construction system must produce contextually rich prompts that shape agent behavior, communication style, and decision-making.
Acceptance Criteria
- Template system for prompt construction (Jinja2 or Python string templates)
- Personality traits affect tone and communication style in generated prompts
- Role description and instructions included in system prompt
- Authority boundaries clearly communicated to the agent
- Company context injected into prompts
- Tool availability described in system prompt
- Token budget awareness (estimate prompt length, stay within limits)
- Unit tests showing different agent configs produce different prompts
- Prompt length estimation utility
- Support for prompt versioning/iteration
Dependencies
- Depends on Implement agent-to-agent messaging with channels and topics #10 (agent engine core)
Design Spec Reference
Section 3.1 — Agent Identity and Personality
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
prio:highImportant, should be prioritizedImportant, should be prioritizedscope:medium1-3 days of work1-3 days of workspec:agent-systemDESIGN_SPEC Section 3 - Agent SystemDESIGN_SPEC Section 3 - Agent Systemtype:featureNew feature implementationNew feature implementation