agentforce-test

Name: agentforce-test
Author: SalesforceAIResearch

179installs37forks

How to install agentforce-test?

salesforceairesearch/agentforce-adlc/agentforce-test

$npx -y skills add SalesforceAIResearch/agentforce-adlc --skill agentforce-test

Installs into the current project.

Summary

Write, run, and analyze structured test suites for Agentforce agents. TRIGGER when: user writes or modifies test spec YAML (AiEvaluationDefinition); runs sf agent test create, run, run-eval, or results commands; asks about test coverage strategy, metric selection, or custom evalu

Files · 1

View on GitHub

SKILL.md

1	# ADLC Test
2
3	Automated testing for Agentforce agents with smoke tests, batch execution, and iterative fix loops.
4
5	## Overview
6
7	This skill provides comprehensive testing capabilities for Agentforce agents, including automated utterance derivation from agent subagents, preview-based smoke testing, trace analysis, and an iterative fix loop for identified issues. It bridges the gap between initial development and production deployment.
8
9	## Platform Notes
10
11	- Shell examples below use bash syntax. On Windows, use PowerShell equivalents or Git Bash.
12	- Replace `python3` with `python` on Windows.
13	- Replace `/tmp/` with `$env:TEMP\` (PowerShell) or `%TEMP%\` (cmd).
14	- Replace `jq` with `python -c "import json,sys; ..."` if jq is not installed.
15	- `find ... \| head -1` -> `Get-ChildItem -Recurse ... \| Select-Object -First 1` in PowerShell.
16
17	## Usage
18
19	This skill uses `sf agent preview` and `sf agent test` CLI commands directly.
20	There is no standalone Python script.
21
22	Quick smoke test (Mode A):
23	```bash
24	# Start preview, send utterance, end session (--authoring-bundle generates local traces)
25	sf agent preview start --json --authoring-bundle MyAgent -o <org-alias>
26	sf agent preview send --json --session-id <ID> --utterance "test" --authoring-bundle MyAgent -o <org-alias>
27	sf agent preview end --json --session-id <ID> --authoring-bundle MyAgent -o <org-alias>
28	```
29
30	Batch testing (Mode B):
31	```bash
32	# Deploy and run test suite
33	sf agent test create --json --spec test-spec.yaml --api-name MySuite -o <org-alias>
34	sf agent test run --json --api-name MySuite --wait 10 --result-format json -o <org-alias>
35	```
36
37	Action execution:
38	```bash
39	# Execute a Flow or Apex action directly via REST API
40	TOKEN=$(sf org display -o <org-alias> --json \| jq -r '.result.accessToken')
41	INSTANCE_URL=$(sf org display -o <org-alias> --json \| jq -r '.result.instanceUrl')
42	curl -s "$INSTANCE_URL/services/data/v63.0/actions/custom/flow/Get_Order_Status" \
43	-H "Authorization: Bearer $TOKEN" -H "Content-Type: application/json" \
44	-d '{"inputs": [{"orderId": "00190000023XXXX"}]}'
45	```
46
47	## Testing Workflow
48
49	This skill supports two testing modes plus direct action execution:
50
51	- Mode A: Ad-Hoc Preview Testing -- Quick smoke tests during development using `sf agent preview`. No test suite deployment needed (org authentication still required). Best for iterative development and fix validation.
52	- Mode B: Testing Center Batch Testing -- Persistent test suites deployed to the org via `sf agent test`. Best for regression suites, CI/CD, and cross-skill integration with /agentforce-observe.
53	- Action Execution -- Direct invocation of Flow/Apex actions via REST API for isolated testing and debugging.
54
55	When to use which:
56
57	\| Scenario \| Mode \|
58	\|----------\|------\|
59	\| Quick smoke test during authoring \| Mode A \|
60	\| Validate a fix from /agentforce-observe \| Mode A \|
61	\| Build a regression suite for CI/CD \| Mode B \|
62	\| Deploy tests to share with the team \| Mode B \|
63	\| Test a single Flow or Apex action in isolation \| Action Execution \|
64
65	---
66
67	## Mode A: Ad-Hoc Preview Testing
68
69	> Full reference: `references/preview-testing.md`
70
71	### Test Case Planning
72
73	If no utterances file is provided, auto-derive test cases from the `.agent` file:
74	1. Subagent-based utterances -- one per non-start subagent from description keywords
75	2. Action-based utterances -- target each key action
76	3. Guardrail test -- off-topic utterance
77	4. Multi-turn scenarios -- subagent transitions
78	5. Safety probes -- adversarial utterances (always included)
79
80	Always present the plan first -- never silently auto-run tests without showing what will be tested. Ask the user to review/modify before executing.
81
82	### Preview Execution
83
84	Use `--authoring-bundle` to compile from the local `.agent` file (enables local trace files):
85
86	```bash
87	SESSION_ID=$(sf agent preview start --json \
88	--authoring-bundle MyAgent \
89	--target-org <org> 2>/dev/null \
90	\| jq -r '.result.sessionId')
91
92	RESPONSE=$(sf agent preview send --json \
93	--session-id

People who liked this also viewed

6 picks

Type

agentforce-test

179installs37forks

View on GitHub

How to install agentforce-test?

salesforceairesearch/agentforce-adlc/agentforce-test

$npx -y skills add SalesforceAIResearch/agentforce-adlc --skill agentforce-test

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# ADLC Test
2
3	Automated testing for Agentforce agents with smoke tests, batch execution, and iterative fix loops.
4
5	## Overview
6
7	This skill provides comprehensive testing capabilities for Agentforce agents, including automated utterance derivation from agent subagents, preview-based smoke testing, trace analysis, and an iterative fix loop for identified issues. It bridges the gap between initial development and production deployment.
8
9	## Platform Notes
10
11	- Shell examples below use bash syntax. On Windows, use PowerShell equivalents or Git Bash.
12	- Replace `python3` with `python` on Windows.
13	- Replace `/tmp/` with `$env:TEMP\` (PowerShell) or `%TEMP%\` (cmd).
14	- Replace `jq` with `python -c "import json,sys; ..."` if jq is not installed.
15	- `find ... \| head -1` -> `Get-ChildItem -Recurse ... \| Select-Object -First 1` in PowerShell.
16
17	## Usage
18
19	This skill uses `sf agent preview` and `sf agent test` CLI commands directly.
20	There is no standalone Python script.
21
22	Quick smoke test (Mode A):
23	```bash
24	# Start preview, send utterance, end session (--authoring-bundle generates local traces)
25	sf agent preview start --json --authoring-bundle MyAgent -o <org-alias>
26	sf agent preview send --json --session-id <ID> --utterance "test" --authoring-bundle MyAgent -o <org-alias>
27	sf agent preview end --json --session-id <ID> --authoring-bundle MyAgent -o <org-alias>
28	```
29
30	Batch testing (Mode B):
31	```bash
32	# Deploy and run test suite
33	sf agent test create --json --spec test-spec.yaml --api-name MySuite -o <org-alias>
34	sf agent test run --json --api-name MySuite --wait 10 --result-format json -o <org-alias>
35	```
36
37	Action execution:
38	```bash
39	# Execute a Flow or Apex action directly via REST API
40	TOKEN=$(sf org display -o <org-alias> --json \| jq -r '.result.accessToken')
41	INSTANCE_URL=$(sf org display -o <org-alias> --json \| jq -r '.result.instanceUrl')
42	curl -s "$INSTANCE_URL/services/data/v63.0/actions/custom/flow/Get_Order_Status" \
43	-H "Authorization: Bearer $TOKEN" -H "Content-Type: application/json" \
44	-d '{"inputs": [{"orderId": "00190000023XXXX"}]}'
45	```
46
47	## Testing Workflow
48
49	This skill supports two testing modes plus direct action execution:
50
51	- Mode A: Ad-Hoc Preview Testing -- Quick smoke tests during development using `sf agent preview`. No test suite deployment needed (org authentication still required). Best for iterative development and fix validation.
52	- Mode B: Testing Center Batch Testing -- Persistent test suites deployed to the org via `sf agent test`. Best for regression suites, CI/CD, and cross-skill integration with /agentforce-observe.
53	- Action Execution -- Direct invocation of Flow/Apex actions via REST API for isolated testing and debugging.
54
55	When to use which:
56
57	\| Scenario \| Mode \|
58	\|----------\|------\|
59	\| Quick smoke test during authoring \| Mode A \|
60	\| Validate a fix from /agentforce-observe \| Mode A \|
61	\| Build a regression suite for CI/CD \| Mode B \|
62	\| Deploy tests to share with the team \| Mode B \|
63	\| Test a single Flow or Apex action in isolation \| Action Execution \|
64
65	---
66
67	## Mode A: Ad-Hoc Preview Testing
68
69	> Full reference: `references/preview-testing.md`
70
71	### Test Case Planning
72
73	If no utterances file is provided, auto-derive test cases from the `.agent` file:
74	1. Subagent-based utterances -- one per non-start subagent from description keywords
75	2. Action-based utterances -- target each key action
76	3. Guardrail test -- off-topic utterance
77	4. Multi-turn scenarios -- subagent transitions
78	5. Safety probes -- adversarial utterances (always included)
79
80	Always present the plan first -- never silently auto-run tests without showing what will be tested. Ask the user to review/modify before executing.
81
82	### Preview Execution
83
84	Use `--authoring-bundle` to compile from the local `.agent` file (enables local trace files):
85
86	```bash
87	SESSION_ID=$(sf agent preview start --json \
88	--authoring-bundle MyAgent \
89	--target-org <org> 2>/dev/null \
90	\| jq -r '.result.sessionId')
91
92	RESPONSE=$(sf agent preview send --json \
93	--session-id

People who liked this also viewed

6 picks

Type