zwright8 Memory Regression Sentinel
Mission
We need this skill because agents lose performance when lessons are not retained and reused. This specific skill prevents unnoticed quality drift after updates.
Activation Cues
- Task requires baseline-delta detection in Memory and Knowledge Operations.
- Task needs explicit risk controls, approval gates, and traceable outcomes.
- Task output must include artifact handoff for humans and agents.
Execution Plan
- Define the scope and success metrics for
Memory Regression Sentinel, including at least three measurable KPIs tied to repeated mistakes and context loss. - Design and version the input/output contract for episodic logs, knowledge nodes, and retrieval metadata, then add schema validation and failure-mode handling.
- Implement the core capability using baseline-delta detection, and produce regression watchlists with deterministic scoring.
- Integrate the skill into swarm orchestration: task routing, approval gates, retry strategy, and rollback controls.
- Add unit, integration, and simulation tests that explicitly cover repeated mistakes and context loss, then run regression baselines.
- Deploy behind a feature flag, monitor telemetry/alerts for two release cycles, and iterate thresholds based on observed outcomes.
Runbook
Preflight:
- None specified.
Execution:
- None specified.
Recovery:
- None specified.
Handoff:
- None specified.
Guardrails
- [quality] Require validations before promoting outputs.
Success Metrics
- Primary metric: repeated mistakes
- Secondary metrics: context loss, decision drift
- Review cadence: weekly
Output Contract
- Return a concise execution summary with key decisions.
- Return risk and mitigation notes with unresolved blockers.
- Return artifact target:
regression watchlists. - Return recommended follow-up tasks for next wave execution.