All agents
OperationsFor teams & operatorsIn build

Process Automation Scout

Finds repetitive work worth automating from run history and process descriptions, estimates frequency and effort in code, and maps each candidate to an existing agent, a config change, or a new-agent proposal.

Automation candidates actioned≥ 25%GoalSign up to run

How it works

Step 1
You provide
Paste your details
Step 2
Grounds
Retrieves & cites — no ungrounded claims
Step 3
Drafts
Schema-checked, code-assembled output
Step 4
You approve
Drafts, you review every step

Ask it like

what should we automate nextfind automation opportunities

Model & guardrails

  • Runs on MYAM shared · open 7B (Qwen) by default — you can point it at your own model.
  • Autonomy: Drafts, you review every step. Nothing leaves without a human’s approval.
  • Data: Internal data. Provenance is attached to every output.

Outcome

Automation candidates actioned
≥ 25%
Goalour goal — not yet measured on live runs
TargetEvalPlatformVerified

Fills in as real usage proves it — from a Target we set, to Eval-measured (passes our test suite), to Platform-measured (across ≥20 accounts), to Verified (customer outcomes).

Also tracked: Annual hours identified · Ceiling-violating proposals = 0

Works well with