outcomes.summaryModule 55Connect governed agent activity to outcomes you can measure, with every modeled figure shown next to the assumption that produced it.
outcomes.summary reports metrics across nine outcome kinds, graded direct, modeled, partial or unconfigured by whether value came from a measured duration or an assumption. builds the full value ledger, with every figure's provenance traced to the event or assumption behind it.
Put admission in front of your MCP servers, repositories, SaaS tools and cloud accounts. Every agent gets a lease or a reason – and you keep the lever to stop any of them.
© 2026 Intramedial. All rights reserved. Intramedial is a product by Trustvice. Secured and supercharged by Avonao Labs.
Report abuse
Plane module
exportwindowassumptionVersionnull if none is set.metrics{ value, unit, confidence, provenance }, with where it applies./docs/outcomes-summaryAPI referenceRoutes, bodies and status codes./docs/outcomes-summaryOpen the consoleSee the module at work in the dashboard, on a seeded fleet or your own./dashboard-preview/impactStart freeCreate a workspace and admit your first intent./auth/signup
currencyconfidencedirect, modeled, partial or unconfigured, naming how a metric's value was produced.provenancecomparisonrecord logs one of nine kinds – agent task completed, conflict prevented, settlement closed and others – with an optional directDurationMinutes when the time was actually measured.
Measured minutes sum only events that supplied a real duration; minutes without one are modeled by multiplying the event count by that kind's assumption, never defaulted to zero.
Labor value comes from measured plus modeled minutes at the configured hourly cost; avoided-incident value comes from a separate per-incident assumption. Either stays unconfigured rather than silently reading as zero if its assumption is missing.
assumptions.set appends a new version rather than replacing the last one, Merkle-receipted over every version so far, and assumptions.history reads the full chain.
Runs in a shared demo workspace with a demo key: anyone can see what it holds, so send nothing private. outcomes.summary docs
Press Run to send the request on the left to the shared demo workspace.
Runs in a shared demo workspace with a demo key: anyone can see what it holds, so send nothing private.
A request in the shape the API accepts, and an example response in the shape the module returns.
GET /api/v1/acp/outcomes/summary Authorization: Bearer av_workspace_•••• → 200 { window, assumptionVersion, metrics, comparison? } (no top-level outcomes or assumptions; the full assumptions object comes from assumptions.get, assumptions.history or export)
{
"window": {
"from": "2026-10-01T00:00:00.000Z",
"to": "2026-11-01T00:00:00.000Z"
},
"assumptionVersion": 3,
"metrics": {
"directCount": {
"value": 212,
"unit": "events",
"confidence": "direct",
"provenance": [
"event:evt_90c1"
]
},
"monetaryValueMinor": {
"value": 184500,
"unit": "minor",
"confidence": "modeled",
"currency": "USD",
: [
]
}
}
}A finance reviewer reads confidence on each metric and knows immediately which figures came from recorded durations and which from an assumption the organization set itself.
An organization lowers its assumed minutes saved per automated task; the change is a new version, Merkle-receipted, and every past summary still points at the version that was actually in force then.
export builds a snapshot with every event, every assumption version, the formulas used and a Merkle root over the whole thing, ready to hand to someone outside the team.
A monetary value stays unconfigured rather than reading as zero when its assumption is missing, so nobody mistakes an unset number for an actual finding of no value.
unconfigured, never a false zero.provenance names the specific events or assumptions it came from.A module is a strong claim about a narrow thing. Rely on it for exactly what it is.
unconfigured, never a default of zero that could be mistaken for a real finding.Agent task completed, conflict prevented, approval resolved, automation completed, evidence pack generated, monitor incident detected, settlement closed, remediation completed and lens finding prevented.
Outcomes has no README; the formulas and field names described here come directly from its service code and its own in-code FORMULAS documentation object.
assumptions.set is idempotent per key; replaying the same key with a different input is refused as a conflict rather than silently applied twice.
unconfigured, not estimated some other way.Authenticate with a workspace API key as a Bearer token. The base URL is https://api.intramedial.com.
/api/v1/acp/outcomes/summaryRead this window's metrics by confidence./api/v1/acp/outcomes/recordLog one outcome event./api/v1/acp/outcomes/exportBuild the full Merkle-receipted value ledger.This module has no MCP action. Call it over REST from any agent or program.
settlement.list47Settlement and escrowHold money against an agent-payment mandate until delivery is confirmed, with a record that says plainly when funds actually moved and when they did not.remediation.blast50Blast map and remediationSee what a declared checkpoint's data touched before you undo anything, then restore against it only once a person confirms it by typed phrase.compliance.pack54Compliance evidence packsSee every control of the regulatory map, with exactly what this workspace can evidence today and what it cannot, kept listed either way.civics.records48Public records searchAsk the record what happened to one resource, or read the workspace's events plainly, redacted to what your permission actually allows.ledger.blame10Ledger, replay and blameTrace every touch to one resource in order, and replay any event to rebuild the receipt that shows it actually happened.ledger-bundles.snapshot49Signed evidence bundlesSeal your workspace's audit chain into a signed bundle an auditor can check offline, against a key published for anyone to read.portmind.memory51