Skip to main content
Scout is the QA Tester of your Agent Army. He reviews pull requests, tests changes against acceptance criteria from Linear tickets, files bugs when something doesn’t pass, and can auto-fix simple build and test failures.

​
What Scout Does

  • Monitors GitHub for PRs ready for review
  • Reviews code changes against Linear ticket acceptance criteria
  • Runs builds and test suites to validate changes
  • Auto-fixes simple build or test failures and retries
  • Files bugs back to Linear with specific gaps when requirements aren’t met
  • Labels PRs as qa-passed, needs-work, or qa-failed
  • Resolves PR review comments by parsing, applying fixes, and pushing

​
Heartbeat

Scout’s heartbeat fires every minute and covers two main jobs:
  1. Resolve review comments — scan all open PRs for unresolved review comments. For each one, use the pr-review-resolver skill to parse comments, apply fixes via Claude Code, and push.
  2. Find PRs to test — query Linear for tickets in “In Review” or “In Review by Agent” with a linked PR. Skip any ticket already tested at the same PR head commit.
  3. Review against acceptance criteria — for each untested ticket, check out the PR branch and review the diff against the Linear ticket’s acceptance criteria. If requirements are unmet, comment on Linear with specific gaps and label needs-work.
  4. Run build and tests — auto-detect the package manager, run install, build, and test suites. On pass: comment on Linear with a checkmark and label qa-passed.
  5. Auto-fix failures — on build or test failure, invoke Claude Code with the error context to apply a minimal fix, commit, push, and re-run once. If still failing after retry, comment on Linear with the error summary and label qa-failed.

​
Skills

SkillPurpose
pr-review-resolverParse PR review comments, apply fixes via Claude Code, and push

​
Workspace Files

Scout’s behavior is defined by workspace files injected into ~/.openclaw/workspace/ during bootstrap:
FileWhat It Does
SOUL.mdCore personality — approach, values, superpowers, boundaries. Defines who Scout is as a QA tester.
IDENTITY.mdName, role label, emoji, avatar metadata.
HEARTBEAT.mdThe state machine logic executed every minute (described above).
TOOLS.mdTool-specific notes and common commands for Scout’s workflow.
He also receives the shared base files (USER.md, AGENTS.md, BOOTSTRAP.md) that all agents share.

​
Configuration

Scout uses the tester preset in your manifest: 
{
  "name": "agent-tester",
  "displayName": "Scout",
  "role": "tester",
  "preset": "tester",
  "volumeSize": 30
}
FieldDefaultNotes
presettesterBuilt-in QA tester preset
volumeSize30 GBPersistent storage — Scout checks out PR branches but doesn’t need as much space
instanceTypeProvider defaultOverride if needed
envVars—Add agent-specific environment variables

​
Customization

You can customize Scout by editing the preset files in the presets/ directory, or by overriding fields in the manifest. To create a completely custom QA agent, set preset: null and provide your own soulContent and identityContent. See Agent Roles & Customization for details.