| name | runtime-parity-audit |
| description | Audit and reconcile drift between agent/prompt docs and actual runtime capabilities in SplitBrain.AI services, controllers, and MCP endpoints. |
| argument-hint | Prompt scope and runtime surfaces to verify |
| user-invocable | true |
| disable-model-invocation | false |
Runtime Parity Audit
Inherits from task-core-loop.
Added Context
Use this to validate that prompt contracts match implemented behavior across the SplitBrain.AI stack.
Focus Surfaces
.github/agents/smith.agent.md
.github/copilot-instructions.md
.github/skills/*/SKILL.md
Docs/Prompts/ if present
- Wiki service (
MemorySmith.App) prompt files under Data/ or Docs/
- Orchestrator API surface (
src/Orchestrator.*/)
- MCP endpoint behavior
Additional Procedure
- Extract behavioral claims from prompts/docs.
- Map each claim to source code evidence.
- Classify each claim as: aligned, drifted, stale, or unverifiable.
- Patch prompt/docs or code based on source-of-truth intent.
- Record explicit drift findings and follow-up tasks.
Output
- Parity matrix (claim -> code evidence -> status).
- Required updates with risk and confidence.