| name | xvision-flywheel-ops |
| description | Operate and audit the xvision self-improvement flywheel across memory capture, autooptimizer, training run example pools, child training, lineage, velocity, and look-ahead-protection evidence. |
xvision flywheel ops
Use this skill when driving the closed loop: capture, observe, score, distill,
optimize, train, gate, activate, recall, and retire.
When to use
Run this skill when auditing or driving the closed improvement loop:
checking flywheel status and velocity, reading lineage, managing training-run
example pools, or gathering weekly/release evidence across the full cycle.
When NOT to use
- Distilling Patterns or running the autoresearch gate → use
xvision-autoresearch-ops
- Listing/inspecting Observations or Patterns or running leakage probes → use
xvision-memory-ops
- Triggering a live trade → no skill applies
Trigger examples
- Show me the flywheel status for agent abc123
- What is the 7-day velocity for agent abc123?
- List the 20 most recent lineage entries for agent abc123
- Run the memory-demos audit script for agent abc123
- Export the flywheel velocity report for the weekly review
- Run the leakage regression and the MCP tests before this release
- Check whether the demo source is frozen-snapshot or fresh-recorder
- What evidence do I need to collect for the flywheel release checklist?
Read Surfaces
xvn flywheel status --agent <agent_id> --json
xvn flywheel velocity --agent <agent_id> --days 7 --json
xvn flywheel lineage --agent <agent_id> --limit 20 --json
Dashboard equivalents live in the memory/flywheel panels. MCP read equivalents:
xvn_flywheel_status
xvn_flywheel_velocity
Training Run Example Discipline
Note: xvn optimize memory-demos has been removed. Use the audit script
directly and manage demo pools through the dashboard Optimizer panel.
bash scripts/audit-memory-demos.sh --target-agent <agent_id> --memory-agent <agent_id>
Use fresh-recorder only when non-reproducibility is intentional and recorded.
Weekly / Release Evidence
bash scripts/export-flywheel-velocity.sh --agent <agent_id> --days 7
bash scripts/leakage-regression.sh
cargo test -p xvision-mcp mcp_
Record which evidence is automated, which is manual, and which requires
live-provider credentials.
Gotchas
fresh-recorder without rationale: Use fresh-recorder only when
non-reproducibility is intentional and recorded. Default to frozen-snapshot
for reproducible demo pools.
Wrong split for untouched period: The untouched split must be 70/15/15.
Non-standard splits invalidate the look-ahead-protection evidence.
Skipping the MCP test suite: cargo test -p xvision-mcp mcp_ is part of
the release evidence checklist. Omitting it leaves MCP surface regressions
undetected.
Confusing velocity with lineage: Velocity is a rate metric (patterns/day,
gate pass rate). Lineage is the audit trail of individual activations and
retirements. Use the right command for the right question.
Running flywheel status during a live cycle: Flywheel reads are safe any
time, but do not trigger optimize or training-run operations inside a live
trading decision process.
Owner
flywheel-ops track (team/contracts/flywheel-ops.md)