A complaint handler’s day is mostly preparation. Read the message, decide whether it is a complaint, notice the sentence about a husband who has died, open the account, find the fee, work out the interest, look up the right article, write the acknowledgement. The decisions, whether to uphold, what to pay, what to say, take a fraction of the time and carry all of the accountability. InsightService puts eight agents on the preparation and leaves every decision with a person.

In short

  • Eight agents, one job each. Six work on a case in order (triage, vulnerability, research, remediation, reply, supervisor); three work across the book (the triage agent over the inbox, the root-cause analyst, the QA reviewer).
  • Each stops at a prepared decision. Accept carries it out in the person’s name, Reject records the disagreement, Ask sends a question back.
  • None of them can send, pay, classify, close or approve. The API refuses those actions from an agent, whatever its level.
  • Autonomy is capped at L2, Ask first, and every agent starts at L1. The numbers come from deterministic engines; a model, when switched on, only explains them, over numbered facts it must cite.

The eight agents

Each agent is defined once, with what it does, what it is grounded in, the one thing it never invents, and the action it may prepare. The catalogue in the product shows the same list.

AgentWhereWhat it preparesIt never invents
Triage agentCase and inboxTreat the case as a complaint (quoting the sentence); open cases for inbox complaints without oneA complaint: each is quoted from a sentence
Vulnerability spotterCaseHow to adapt the case, from the indicators and the knowledge baseA vulnerability: each indicator is quoted
Case researcherCaseRoot-cause tags, after reading accounts, fees, related cases, prior complaints and fee runsA charge: every fee is the firm’s record
Remediation calculatorCaseThe worked-out remediation: refunds plus simple interestAn amount: the calculator computes it
Reply drafterCaseThe acknowledgement, response or final response, heldA promise: referral wording is a marked template
SupervisorCaseAn outcome where the records support one, and what still blocks the closeA decision: people close cases
Root-cause analystOversightOwnership of the clusters that look systemicA cause: every cluster counts real complaints
QA reviewerOversightPasses for the sampled cases whose checks all passedA pass: a person records each result
The agent catalogue: eight agents across Identify, Investigate and put right, Respond and close, and Oversight, each with its summary, the action it may prepare, its level and what it never invents
The catalogue. Eight agents in four phases, each with the action it may prepare and the line it doesn’t cross.

A run on a case

“Run the agents” on a case runs the six case agents in order on the shared agent pipeline, each as itself and at its own autonomy level. Each step reads what the engines computed, writes a short finding with its evidence, and, if there is something for a person to do, prepares it.

SupervisorReply drafterRemediationResearchVulnerabilityTriageHandlerSupervisorReply drafterRemediationResearchVulnerabilityTriageHandlerRun the agents1Prepares: treat as a complaint (quote)2Prepares: how to adapt the case3Prepares: root-cause tags4Prepares: the remediation5Prepares: the reply, held6Prepares: an outcome, and what blocks the close7

Six case agents in order. Each one prepares; none of them acts on the case without the handler.

On Eilidh Ferreira’s complaint about a Premier fee she kept paying after downgrading, the last run shows what each agent found: the triage agent quoted her sentence about being charged “the £32 monthly fee every single month since”; the vulnerability spotter found no indicators in her words or her record; the case researcher found nine fees in question, £266.00; the remediation calculator’s £273.49 had been submitted for a team lead because it is above the threshold; the reply drafter said no reply was due until the outcome; and the supervisor listed three things standing between the case and its close, then prepared “Uphold SVC-2026-000013: the records show the fees in question” for the handler to accept, reject or ask about.

The Agents tab on SVC-2026-000013, Eilidh Ferreira: one decision waiting, the supervisor’s proposal to uphold with each fee listed and Accept, Reject and Ask buttons; and the last run, with the triage agent, vulnerability spotter, case researcher, remediation calculator, reply drafter and supervisor, each at L1 with its finding
Waiting on the handler. The supervisor’s proposal to uphold, with the evidence, beside the last run of all six case agents.

Across the book: the inbox, root causes and QA

Three runs work on the whole book rather than one case. Over the inbox, the triage agent classifies new contacts and prepares cases for the complaints that don’t have one yet, most urgent first; on the demo there are 1,094 classified complaints without a case, which is exactly the backlog a person should not have to sort by hand. The root-cause analyst clusters every complaint and prepares ownership of the ones that look systemic. The QA reviewer samples closed cases and prepares passes for the clean ones. In each, the pattern is the same as on a case: the engine computes, the agent prepares one decision, and a person answers it.

Accept, reject or ask

A prepared decision appears on the case and on the person’s home page under “What needs me”. There are three answers:

AnswerWhat happens
AcceptThe action is carried out in the person’s name: the case becomes a complaint, the tags are saved, the remediation is proposed, the reply is put on the case held, the outcome is recorded
RejectNothing changes on the case; the disagreement is recorded and counts against the agent’s agreement rate
AskA question goes back; the decision stays open

This is the important detail: accepting a prepared outcome records the outcome as the person’s. The audit trail says who accepted, from which agent, and when. Whether a case is a complaint is a person’s call “the triage agent only proposes it”; the outcome is a person’s call “the supervisor only proposes it”. Those are the API’s own refusal messages when an agent tries.

The decisions no agent can take

ActionWho can do itThe API’s answer to an agent
Classify a contact as a complaint or notA person“A person corrects the classification.”
Send a reply or a request to the customerA person“An agent can’t send to a customer; a person sends it.”
Approve or pay a remediationA person (a team lead above the threshold)“An agent can’t approve or pay a remediation; a person does.”
Record the outcomeA person“The outcome is a person’s call; the supervisor only proposes it.”
Close the caseA person (a team lead when referred)“Closing a case is a person’s call.”
Own a root cause, record a QA resultA personRefused

These refusals don’t depend on the autonomy level. Even at the highest level the platform defines, an agent couldn’t do them in InsightService, because the operations themselves check who is acting.

Autonomy is earned, and capped

The shared agent platform has four levels. InsightService uses the first two.

LevelNameIn InsightService
L1DraftWhere every agent starts: it prepares, a person acts
L2Ask firstThe ceiling for every agent action (a hard cap in the code), and the tenant ceiling via SERVICE_AGENTS_MAX_LEVEL
L3Act and tellNot available
L4ActNot available

90% agreement over 50 answers in 30 days

below 70% over 20 answers

hard cap

L1 Draft

L2 Ask first

L3 and L4: not available

The platform’s default thresholds for recommending a change of level. An administrator makes the change; the cap holds regardless.

Every answer a person gives to a prepared decision is counted. The autonomy page shows each agent’s agreement and override rates over 30 days, per action. The platform’s defaults recommend raising an agent a level only when people agreed with at least 90% of at least 50 answers in that window, and lowering it when agreement falls below 70% over at least 20. A recommendation is just that: an administrator decides, the reason is recorded, and the hard cap at L2 means no recommendation can take an InsightService agent past Ask first.

The Autonomy page: the tenant ceiling at L2, the four levels L1 Draft to L4 Act, and for each agent and action its level, set by whom, and its agreement statistics
Autonomy per agent and action. Every agent at L1 in the demo, under a tenant ceiling of L2, with the agreement record that would justify a change.

The model explains; it never supplies the facts

Every classification, deadline, match and amount in InsightService comes from a deterministic engine: the classifier’s lexicons, the clock, case research over the firm’s records, the remediation calculator, the case controls. The agents run those engines. Where a language model is switched on (SERVICE_AGENTS_LLM), it receives the engine’s results as numbered facts, [F1], [F2] and so on, with an instruction to state nothing that isn’t in them and to cite what it relies on. An answer that carries a name or a number not in the facts is thrown away.

Model off is a supported mode

With the model switched off, as in the demo (SERVICE_AGENTS_LLM=0), every agent runs on its rules and says so: the case shows “Reasoned by rules” on each prepared decision and “model off — rules only” on the run. The product works the same; the explanations are shorter.

The copilot reads, and only reads

Beside the agents sits a copilot for the person working the case. It answers questions about the open case through read-only tools: the summary and clock, the contacts, the customer’s accounts and fees, the research and related cases, the knowledge base, the remediation, the replies, what the agents prepared and the root causes. It never sends, pays, classifies or closes anything. With the model off it answers from the case’s headline facts, and says so.

The InsightService copilot on SVC-2026-000013, with AI off, answering what still blocks closing the case from the case’s headline facts: on track, final response due 2027-01-06 (illustrative policy), nine fees in question, and the three blockers
The copilot with the model off. It reads the case’s facts directly and says that is what it did.

Everything on the record

The activity log lists every agent step on every case and in oversight: when it ran, at what level, what it found, what it prepared and whether that is still waiting on someone. The case’s own history interleaves agent steps and people’s actions, so a reviewer can see, line by line, that the supervisor proposed an outcome and the handler accepted it, or that the remediation calculator prepared £20.06 and the handler proposed it.

Agent activity: 100 steps across cases and the oversight runs, with each step’s agent, subject, finding, status and how long ago, and the counts of prepared, answered and waiting decisions
Agent activity. Every step, what it prepared, and what is waiting on a person.

For the head of complaints

The decisions that carry regulatory accountability, classification, outcome, redress approval, sending and closing, are a person’s by construction, not by policy.

The record shows which agent prepared what and who accepted it.

For the service operations lead

Handlers start every case with the research done, the remediation worked out and the letter drafted.

Agreement rates show which agents the team trusts and which need work.

For the board

Automation that can’t send, pay or close on its own is automation the firm can explain to a regulator and an ombudsman.

Autonomy only grows with evidence, and never past Ask first.

Let the agents do the preparing. Keep the deciding.

Eight agents read, research, calculate and draft. They stop at a prepared decision every time.

People classify, approve, send and close, and the record says so.

Keep reading