utun2/.opencode/context/openagents-repo/lookup/subagent-test-commands.md

Subagent Testing Commands - Quick Reference

Purpose: Quick command reference for testing subagents

Last Updated: 2026-01-07

---

Standalone Mode (Unit Testing)

Run All Standalone Tests

cd evals/framework
npm run eval:sdk -- --subagent=ContextScout --pattern="standalone/*.yaml"

Run Single Test

npm run eval:sdk -- --subagent=ContextScout --pattern="standalone/01-simple-discovery.yaml"

Debug Mode

npm run eval:sdk -- --subagent=ContextScout --pattern="standalone/*.yaml" --debug

---

Delegation Mode (Integration Testing)

Run Delegation Tests

npm run eval:sdk -- --agent=core/openagent --pattern="delegation/*.yaml"

Test Specific Delegation

npm run eval:sdk -- --agent=core/openagent --pattern="delegation/01-contextscout-delegation.yaml"

---

Verification Commands

Check Agent File

# View agent frontmatter
head -30 .opencode/agent/ContextScout.md

# Check tool permissions
grep -A 10 "^tools:" .opencode/agent/ContextScout.md

Check Test Config

cat evals/agents/ContextScout/config/config.yaml

View Latest Results

# Summary
cat evals/results/latest.json | jq '.summary'

# Agent loaded
cat evals/results/latest.json | jq '.meta.agent'

# Tool calls
cat evals/results/latest.json | jq '.tests[0]' | grep -A 5 "Tool"

# Violations
cat evals/results/latest.json | jq '.tests[0].violations'

---

Common Test Patterns

Smoke Test

npm run eval:sdk -- --subagent=ContextScout --pattern="smoke-test.yaml"

Specific Test Suite

npm run eval:sdk -- --subagent=ContextScout --pattern="discovery/*.yaml"

All Tests for Subagent

npm run eval:sdk -- --subagent=ContextScout

---

Flag Reference

Flag	Purpose	Example
--subagent	Test subagent in standalone mode	--subagent=ContextScout
--agent	Test primary agent (or delegation)	--agent=core/openagent
--pattern	Filter test files	--pattern="standalone/*.yaml"
--debug	Show detailed output	--debug
--timeout	Override timeout	--timeout=120000

---

Troubleshooting Commands

Check Which Agent Ran

# Should show subagent name for standalone mode
cat evals/results/latest.json | jq '.meta.agent'

Check Tool Usage

# Should show tool calls > 0
cat evals/results/latest.json | jq '.tests[0]' | grep "Tool Calls"

View Test Timeline

# See full conversation
cat evals/results/history/2026-01/07-*.json | jq '.tests[0].timeline'

Check for Errors

# View violations
cat evals/results/latest.json | jq '.tests[0].violations.details'

---

File Locations

Agent Files

.opencode/agent/subagents/core/{subagent}.md

Test Files

evals/agents/subagents/core/{subagent}/
├── config/config.yaml
└── tests/
    ├── standalone/
    │   ├── 01-simple-discovery.yaml
    │   └── 02-advanced-test.yaml
    └── delegation/
        └── 01-delegation-test.yaml

Results

evals/results/
├── latest.json                    # Latest test run
└── history/2026-01/              # Historical results
    └── 07-HHMMSS-{agent}.json

---

Quick Checks

Is Agent Loaded Correctly?

# Should show: "agent": "ContextScout"
cat evals/results/latest.json | jq '.meta.agent'

Did Agent Use Tools?

# Should show: Tool Calls: 1 (or more)
cat evals/results/latest.json | jq '.tests[0]' | grep "Tool Calls"

Did Test Pass?

# Should show: "passed": 1, "failed": 0
cat evals/results/latest.json | jq '.summary'

---

Related

• concepts/subagent-testing-modes.md - Understand testing modes
• guides/testing-subagents.md - Step-by-step testing guide
• errors/tool-permission-errors.md - Fix common issues

Reference: evals/framework/src/sdk/run-sdk-tests.ts