Analyze delayed management feedback stability

What it's for

Adds control-theoretic organizational intelligence unavailable in engineering dashboards: Gitrevio can reveal when delayed pressure, WIP, quality, incident, and staffing responses create self-amplifying dynamics before leaders intensify the wrong lever.

What you give it

Inputs split into evidence read from your connected systems, calibration your team owns, and numerical controls that affect precision but never the result's meaning.

{
  "damping_test_fraction": 0.25,
  "edges": [
    {
      "coefficient_lower": 1.1,
      "coefficient_upper": 1.2,
      "id": "demand-to-wip",
      "lag_steps": 1,
      "source_id": "demand",
      "target_id": "work_in_progress"
    },
    {
      "coefficient_lower": 1.1,
      "coefficient_upper": 1.2,
      "id": "wip-to-demand",
      "lag_steps": 1,
      "source_id": "work_in_progress",
      "target_id": "demand"
    }
  ],
  "maximum_acceptable_spectral_radius": 0.95,
  "nodes": [
    {
      "id": "demand"
    },
    {
      "id": "work_in_progress"
    }
  ],
  "scenario_draws": 100,
  "seed": 92,
  "transient_horizon": 20
}

What you get back

This is the actual output of running the example above — computed by the same function the platform calls, not an illustration.

{
  "assumptions": [
    "The signed edge coefficients and delays represent a locally linear causal or structural response model at the declared cadence and operating point.",
    "Coefficient intervals cover plausible simultaneous values; sampled corners and interior draws are stress scenarios, not a probability distribution or exhaustive interval proof.",
    "Unmodeled saturation, thresholds, exogenous forcing, control actions, and regime changes are either negligible near the operating point or tested separately."
  ],
  "decision": "delayed_feedback_instability_risk_material",
  "edge_damping_leverage": [
    {
      "edge_id": "demand-to-wip",
      "lag_steps": 1,
      "source_id": "demand",
      "spectral_radius_reduction_under_test_damping": 0.1608,
      "target_id": "work_in_progress",
      "worst_scenario_coefficient": 1.2
    },
    {
      "edge_id": "wip-to-demand",
      "lag_steps": 1,
      "source_id": "work_in_progress",
      "spectral_radius_reduction_under_test_damping": 0.1608,
      "target_id": "demand",
      "worst_scenario_coefficient": 1.2
    }
  ],
  "executive_summary": {
    "fraction_sampled_scenarios_above_governed_limit": 1,
    "fraction_sampled_scenarios_above_one": 1,
    "maximum_transient_amplification": 38.3376,
    "maximum_transient_amplification_step": 20,
    "midpoint_spectral_radius": 1.15,
    "stability_margin_to_one": -0.2,
    "worst_sampled_spectral_radius": 1.2
  },
  "limitations": [
    "Sampled spectral stability is not a formal robust-stability certificate over every point in a continuous coefficient box.",
    "Damping leverage is a one-edge local stress test, not proof that changing a KPI, policy, or organizational relationship will cause the predicted stabilization.",
    "Do not interpret a modeled node or edge as individual intent, blame, performance, or grounds for employment or security action."
  ],
  "method": "interval_stress_delayed_var_companion_stability_v1",
  "sample": {
    "edges": 2,
    "maximum_lag": 1,
    "nodes": 2

How it works

Decision analysis — Turn uncertainty, cost and risk appetite into a defensible choice, with the reasoning left inspectable.

1. 1 Freeze nodes, signed directed effects, coefficient intervals, observation/action cadence, and integer response delays from a causal or explicitly structural system model; reject unknown endpoints, inverted intervals, and oversized delayed state spaces.
2. 2 Construct the delayed VAR companion matrix, evaluate midpoint plus all-lower, all-upper, and reproducible interior coefficient scenarios, and compute each spectral radius to identify sampled asymptotic instability against the governed margin.
3. 3 For the worst sampled matrix, propagate powers over the transient horizon to expose non-asymptotic amplification, then damp each edge toward zero by the same test fraction and recompute spectral radius as local leverage evidence.
4. 4 Report worst/midpoint radii, margin, scenario exceedance fractions, transient amplification, and ranked damping candidates while labeling sampled interval stress as neither probability nor exhaustive robust stability proof.

Before you trust it

Every tool in the catalog ships with the conditions under which its answer is meaningful — and the conditions under which it should abstain instead of guessing.

Calibrating it to your org

Calibration workflow

1. 1 Define the management decision, target outcome, aggregate unit, privacy boundary, cadence, and prediction/intervention horizon for this organization.
2. 2 Build a tenant-scoped historical cohort using only information available before each prediction or decision; preserve zero periods, censoring, assignment probabilities, and unresolved outcomes when the method requires them.
3. 3 Estimate statistical parameters on training history, but obtain costs, utilities, risk tolerance, practical-effect thresholds, capacity, and policy constraints from accountable decision owners.
4. 4 Validate on later time windows or held-out aggregate units at the deployment grain, against a simple baseline and the function-specific validation strategy.
5. 5 Deploy only if the returned decision clears evidence, overlap, calibration, robustness, and guardrail checks; warning, unsupported, schema-gap, and fallback decisions are abstentions.
6. 6 Monitor realized outcomes, data drift, coverage, and decision regret; recalibrate at a governed cadence or after a detected regime/definition change, never merely because a stakeholder dislikes the result.

Call it from your AI

You don't wire up 388 tools in your MCP client. The GitRevio MCP server exposes 18 tools, three of which let an agent search the catalog, read a tool's schema, and run it — so the assistant finds this one on its own.

gitrevio_capabilities_search
  { "q": "stress the dynamic stability of a" }
  → finds "analyze_delayed_management_feedback_stability"

gitrevio_capability_describe
  { "capability_id": "analyze_delayed_management_feedback_stability" }
  → returns the input schema and agent guidance shown on this page

gitrevio_capability_run
  { "capability_id": "analyze_delayed_management_feedback_stability", "arguments": { ... } }
  → returns the result shown above

Works in Claude Desktop, Claude Code, Cursor, Cline, Continue.dev, Goose and Aider. See the MCP server.

Related tools

See every tool in Org design, incentives & decisions →