sf-ai-agentforce-testing（SF AI AgentForce 测试）

sf-ai-agentforce-testing: Agentforce Test Execution & Coverage Analysis

Use this skill when the user needs formal Agentforce testing: multi-turn conversation validation, CLI Testing Center specs, topic/action coverage analysis, preview checks, or a structured test-fix loop after publish.

When This Skill Owns the Task

Use sf-ai-agentforce-testing when the work involves:

• sf agent test workflows
• multi-turn Agent Runtime API testing
• topic routing, action invocation, context preservation, guardrail, or escalation validation
• test-spec generation and coverage analysis
• post-publish / post-activate test-fix loops

Delegate elsewhere when the user is:

• building or editing the agent itself → sf-ai-agentforce or sf-ai-agentscript
• running Apex unit tests → sf-testing
• creating seed data for actions → sf-data
• analyzing session telemetry / STDM traces → sf-ai-agentforce-observability

Core Operating Rules

• Testing comes after deploy / publish / activate.
• Use multi-turn API testing as the primary path when conversation continuity matters.
• Use CLI Testing Center as the secondary path for single-utterance and org-supported test-center workflows.
• Interactive and programmatic CLI preview use standard sf org login web authentication; ECA is only required for Agent Runtime API testing, not for live preview.
• Fixes to the agent should be delegated to sf-ai-agentscript when Agent Script changes are needed.
• Do not use raw curl for OAuth token validation in the ECA flow; use the provided credential tooling.

Script path rule

Use the existing scripts under:

• ~/.claude/skills/sf-ai-agentforce-testing/hooks/scripts/

These scripts are pre-approved. Do not recreate them.

Required Context to Gather First

Ask for or infer:

• agent API name / developer name
• target org alias
• testing goal: smoke test, regression, coverage expansion, or bug reproduction
• whether the agent is already published and activated
• whether the org has Agent Testing Center available
• whether ECA credentials are available for Agent Runtime API testing

Preflight checks:

1. discover the agent
2. confirm publish / activation state
3. verify dependencies (Flows, Apex, data)
4. choose testing track

Dual-Track Workflow

Track A — Multi-turn API testing (primary)

Use when you need:

• multi-turn conversation testing
• topic re-matching validation
• context preservation checks
• escalation or action-chain analysis across turns

Requires:

• ECA / auth setup
• agent runtime access

Track B — CLI Testing Center (secondary)

Use when you need:

• org-native sf agent test workflows
• test spec YAML execution
• quick single-utterance validation
• CLI-centered CI/CD usage where Testing Center is available

Quick manual path

For manual validation without full formal testing, use preview workflows first, then escalate to Track A or B as needed.

Recommended Workflow

1. Discover and verify

• locate the agent in the target org
• confirm it is published and activated
• confirm required actions / Flows / Apex exist
• decide whether Track A or Track B fits the request

2. Plan tests

Cover at least:

• main topics
• expected actions
• guardrails / off-topic handling
• escalation behavior
• phrasing variation

3. Execute the right track

Track A

• validate ECA credentials with the provided tooling
• retrieve metadata needed for scenario generation
• run multi-turn scenarios with the provided Python scripts
• analyze per-turn failures and coverage

Track B

• generate or refine a flat YAML test spec
• run sf agent test commands
• inspect structured results and verbose action output

4. Classify failures

Typical failure buckets:

• topic not matched
• wrong topic matched
• action not invoked
• wrong action selected
• action invocation failed
• context preservation failure
• guardrail failure
• escalation failure

5. Run fix loop

When failures imply agent-authoring issues:

• delegate fixes to sf-ai-agentscript
• re-publish / re-activate if needed
• re-run focused tests before full regression

Testing Guardrails

Never skip these:

• test only after publish/activate
• include harmful / off-topic / refusal scenarios
• use multiple phrasings per important topic
• clean up sessions after API tests
• keep swarm execution small and controlled

Avoid these anti-patterns:

• testing unpublished agents
• treating one happy-path utterance as coverage
• storing ECA secrets in repo files
• debugging auth with brittle shell-expanded curl commands
• changing both tests and agent simultaneously without isolating the cause

Output Format

When finishing a run, report in this order:

1. Test track used
2. What was executed
3. Pass/fail summary
4. Coverage gaps
5. Root-cause themes
6. Recommended fix loop / next test step

Suggested shape:

Agent: <name>
Track: Multi-turn API | CLI Testing Center | Preview
Executed: <specs / scenarios / turns>
Result: <passed / partial / failed>
Coverage: <topics, actions, guardrails, context>
Issues: <highest-signal failures>
Next step: <fix, republish, rerun, or expand coverage>

Cross-Skill Integration

Reference Map

Start here

• references/interview-wizard.md
• references/multi-turn-testing.md
• references/cli-commands.md
• references/test-spec-reference.md

Execution / auth

• references/execution-protocol.md
• references/multi-turn-execution.md
• references/eca-setup-guide.md
• references/credential-convention.md
• references/connected-app-setup.md

Coverage / fix loops

• references/coverage-analysis.md
• references/agentic-fix-loops.md
• references/results-scoring.md
• references/known-issues.md

Advanced / specialized

• references/agentscript-agents.md
• references/agentscript-testing-patterns.md
• references/cli-testing-details.md
• references/deep-conversation-history-patterns.md
• references/swarm-execution.md
• references/trace-analysis.md
• references/agent-api-reference.md

Templates / assets

• references/test-templates.md
• references/test-plan-format.md
• assets/

Score Guide