Files
fidelity-ai-workspace/agent-memory/behavior/learning-sessions.md
david.delagneau dbc1894e27 Add project-knowledge structure and templates
- Introduced new maps for navigating project knowledge, including "Current Work," "Fidelity Domain," "Fidelity Apps," "Work Items," and "People."
- Created base files for daily notes, decisions, people, systems, work items, and workstreams with defined properties and views.
- Developed templates for daily notes, decisions, meeting notes, persons, systems, work items, and workstreams to standardize documentation.
- Updated scripts and prompts to reflect the new project-knowledge directory structure.
- Removed outdated onboarding and start-here documents, consolidating relevant information into the new maps.
- Ensured all references in workflows and scripts point to the new project-knowledge paths.
2026-04-17 15:52:08 -06:00

1.2 KiB

type, status, updated, tags
type status updated tags
agent-behavior active 2026-04-17
agent
learning

Learning Sessions

Learning sessions are used to clarify durable project understanding, not to chase transient ticket status.


Rules

  • Answer from known context and verified facts only.
  • Label unknowns, assumptions, and inferences.
  • Ask 3 to 5 high-leverage questions when the user asks what should be clarified.
  • Prefer questions about architecture, ownership, process, release mechanics, debugging strategy, domain concepts, and decision rules.
  • Avoid questions that only ask whether a task moved today unless the answer changes a durable rule.
  • When the user teaches or corrects the agent, update the smallest correct canonical file or behavior surface.

Good Clarification Targets

  • Ownership boundary between app, SDK, adapter, feature module, backend service, and backend-driven configuration.
  • Release propagation path across SDK, adapter, consuming app, and validation environments.
  • Evidence required before classifying an external issue as a regression.
  • REST vs GraphQL parity rules and fallback behavior.
  • Authenticated vs non-authenticated flow differences.