Agentforce for ITSM Setup Orchestrator

Guide the user through setting up Agentforce Studio, the IT Service Fulfiller agent, the IT Service Employee agent, specialized employee agents, and Employee-Agent to human escalation in Salesforce Service Cloud ITSM by presenting the available capabilities, delegating to specialized child skills, and tracking progress.

Goal

Act as the coordinator for Agentforce feature configuration in ITSM. Present the user with a menu of configurable features, invoke the appropriate child skill for each selection, and after each feature completes, return to the menu with updated progress until the user is done.

Behavior

1. Extract context from conversation

Before presenting options, scan chat history for:

• Which features the user has already set up (skip or mark as done)
• Any preferences or constraints mentioned (e.g., "just enable Agentforce Studio", "we already have Studio on")
• The target org (if mentioned)
• Business context that informs which features are relevant

2. Confirm the target org

Agentforce setup performs writes against a real org (feature-toggle enablement, agent creation and activation). Before delegating to any child skill, confirm the target org with the user and state plainly that this org will be modified. Never assume production is safe to change — ask for explicit confirmation of the org.

3. Present the Agentforce feature menu as a multi-select

Show the user what's available and what's done, organized into the two sequential setup stages — Stage 1: enable platform features (Agentforce Studio enablement) and Stage 2: install & activate agent templates (the Fulfiller and Employee agents). This separation is load-bearing: Stage 1 enables org-level platform toggles/preferences, while Stage 2 installs and activates agents from a template — they are distinct kinds of action and must read as distinct stages, not one undifferentiated list. Only features with a working child skill appear in the menu — use the Feature menu template in examples/output-templates.md. Collect the user's selections through a single multi-select prompt (use AskUserQuestion with multiSelect: true when tooling permits, otherwise ask the user to reply with a list of numbers such as 1, 2). Use the exact Item names from the Feature menu table as the multi-select option labels — verbatim, including each item's parenthetical note and the word "Agent" and its capitalization (Agentforce Studio enablement (Foundation for all agents), IT Service Fulfiller Agent, IT Service Employee Agent, Specialized Agents for Employee, Employee Agent escalation) — so the picker options and the table never diverge. Each item's Description column matches the wording on the Salesforce Setup → Agentforce for IT Service page (the Fulfiller and Employee agent-template descriptions and the Specialized Agents for Employee text are used verbatim) so a user evaluating which template to install reads the same purpose/scope here as they would in Setup — do not paraphrase or shorten that product copy. For the three Stage 2 agent items, follow the verbatim product copy with one short trailing Setup: line stating what installing does (Setup: creates the agent from this template and activates a version. for the Fulfiller and Employee agents; Setup: pick one specialized template (e.g. Password Manager, Onboarding), then create and activate that standalone agent. Re-run this item to add more. for Specialized Agents for Employee, for which you first ask the user which specialized template they want and pass that name to the child skill) so a single row carries purpose, scope, and the install action together. The Stage 1 row (Agentforce Studio enablement) is a platform-enablement toggle, not a template install, so it carries no Setup: line. When raising AskUserQuestion, put that same Description text (product copy + the Setup: line) in each option's description field so the picker carries the detail too. Do NOT show placeholder features that cannot be executed.

Report file (harness / non-interactive runs). If a ${outputDir} is provided (via the harness's generated-file location directive), write the menu emission (attribution header + feature table with status + delegation targets + dependency signal + the multi-select prompt itself) to ${outputDir}/report.md before raising AskUserQuestion — so the report file always exists even when the harness parks at the confirmation gate. Overwrite the same file after each feature completes with the updated status table. Skip these writes when running interactively for a user in a chat surface — write only when ${outputDir} was passed as an explicit destination.

4. Delegate to child skills in dependency order

Studio-first rule (unconditional). If Agentforce Studio enablement (#1) is in the user's selection and not already done, run it first, always — regardless of the order the user listed their numbers in. The Fulfiller Agent (#2), Employee Agent (#3), and Specialized Agents for Employee (#4) all depend on Studio being enabled and will fail if attempted first. Reorder the queue silently so Studio runs before any agent. This rule is non-negotiable and applies whether the user selected two features (Studio + one agent) or everything. Employee Agent escalation (#5) is a post-setup action that also never runs before Studio.

User-order rule (among #2, #3, and #4 only). The Fulfiller Agent (#2), Employee Agent (#3), and Specialized Agents for Employee (#4) are independent of each other — none depends on the others. If more than one is selected, run them in the order the user listed them (default 2 → 3 → 4 when unspecified). This rule applies only to the ordering among #2, #3, and #4; it never overrides the Studio-first rule above.

Escalation-after-Employee rule (#5 depends on #3). Employee Agent escalation (#5) configures the Employee agent's hand-off to a human, so it requires the Employee Agent (#3) to exist and be Active first. If #5 is selected, ensure #3 has completed successfully in this run (or was already done) before delegating to it; otherwise queue #3 ahead of #5. #5 never runs before Studio (#1) either.

Runtime-access-after-creation rule (unconditional). Whenever any Stage 2 agent (#2, #3, #4) returns a live verdict this session (CREATED/ALREADY-CREATED/ACTIVATED), append an automatic Stage 3 runtime-access step at the back of the queue — the post-requisite mirror of the Studio-first pre-requisite: appended always, never a menu pick — and delegate it once after the last Stage 2 agent to service-itsm-agentic-setup-agent-runtime-access-assign, covering every newly-live agent (its own target-user + confirm-to-write gates govern the grants; Stage 3 guarantees the offer, never a silent grant). It is non-negotiable: a live agent's actions fail on missing user permissions until this sets them up. Do NOT append it when only Studio ran or no agent went live (FAILED/PARTIAL/PENDING CONFIRMATION/DECLINED, or a Stage 2 verify that stopped the queue). Narration + status mapping: Runtime access hand-off in examples/output-templates.md.

#	Feature	Stage	Child Skill
1	Agentforce Studio enablement (Foundation for all agents)	1	service-itsm-agentic-setup-agentforce-studio-configure
2	IT Service Fulfiller Agent	2	service-itsm-agentic-setup-fulfiller-agent-configure
3	IT Service Employee Agent	2	service-itsm-agentic-setup-employee-agent-configure
4	Specialized Agents for Employee	2	service-itsm-agentic-setup-employee-agent-configure
5	Employee Agent escalation	Post	service-agentforce-human-escalation-configure (pass the IT scenario inputs: agent IT_Service_Employee_Agent, queue General_IT_Queue, CONTEXT_OBJECT=MessagingSession)
—	Runtime access (auto after any Stage 2 create/activate; never a user pick)	3	service-itsm-agentic-setup-agent-runtime-access-assign

IT Service Employee Agent vs Specialized Agents for Employee. Both delegate to the same child skill, service-itsm-agentic-setup-employee-agent-configure — the difference is which template it installs. The IT Service Employee Agent installs the broad, ready-to-go employee agent (the child skill's default when no specialization is named). Specialized Agents for Employee installs a specialized one instead — but the child skill only takes the specialized path when handed a specialized template name; handed nothing it silently falls back to the broad agent, re-creating the IT Service Employee Agent. So before delegating Specialized Agents for Employee, ask which specialized employee agent the user wants — offer common examples (Password Manager, Certificate Management, Onboarding, Hardware Request), note more are available, and let the child skill disambiguate a partial or ambiguous name; then delegate with that named specialization (the chosen template names the agent). Never delegate Specialized Agents for Employee without a named specialization — that is the one case that produces a duplicate broad agent. The specialized templates themselves are turned on in Stage 1; this item is where an agent is built and activated from one of them.

service-itsm-agentic-setup-agentforce-studio-configure performs its own read-and-classify preflight (reading live toggle state before writing) rather than delegating to service-itsm-agentic-setup-agentforce-studio-validate — that skill is a separate, read-only entry point a user can invoke directly to check readiness without writes. This orchestrator does not need to call it as part of the delegation flow above.

5. After each feature completes

Once a child skill finishes:

1. Verify the child skill's own deterministic verdict by running node "<skill_dir>/scripts/verify-child-verdict.mjs" <studio|fulfiller|employee|escalation|runtime> <verdict> — never re-derive the success/failure comparison in prose. Pass Studio's overall field from classify-final-report.mjs, Fulfiller/Employee Agent's Phase 8 aggregate verdict, the escalation leaf's status (CONFIGURED/ALREADY-CONFIGURED), or the runtime-access skill's Phase-7 aggregate, as <verdict>. Exit code 0 means advance; exit code 1 means stop and surface the failure in plain language — do not advance to the next feature in the queue. A partially-enabled Studio (e.g. Einstein GenAI on but the parent umbrella still blocked, overall: PARTIAL) will make Fulfiller/Employee Agent creation fail too, so the script treats PARTIAL the same as FAILED for advancement purposes.
2. Update the status — mark the completed feature as "Done"
3. Suggest the next logical step — if another feature is available, recommend it based on the dependency order
4. Re-present the menu with updated status — use the Post-feature progress template in examples/output-templates.md
5. Stage 3 after the Stage 2 queue drains. Once no Stage 2 agent remains queued and at least one went live this session, delegate the appended Stage 3 once (per the rule above) and verify with verify-child-verdict.mjs runtime <verdict> — exit 0 ⇒ Stage 3 Done → completion summary; exit 1 ⇒ stop and surface in plain language.

6. Completion summary

When the user says they're done (or all available features are configured), present a final summary using the Completion summary template in examples/output-templates.md.

---

Setup Stages & Recommended Order

Setup runs in two sequential setup stages, followed by optional post-setup escalation. Stage 1 (enable platform features) must complete before Stage 2 (install & activate agent templates). Escalation runs only after the Employee agent is active.

Stage 1 — Foundation: enable platform features
  1. Agentforce Studio enablement (Foundation for all agents)   (turn ON org-level Agentforce + Einstein GenAI feature toggles)

Stage 2 — Agent templates: install & activate   (only after Stage 1)
  2. IT Service Fulfiller Agent          (install from template, commit, and activate the agent)
  3. IT Service Employee Agent           (install the broad employee agent from its template and activate it)
  4. Specialized Agents for Employee     (install a specialized employee agent — user picks the template — and activate it)

Stage 3 — Runtime access   (AUTO — after any Stage 2 agent goes live, not a menu pick)
  •  Grant the live agent(s)' runtime feature permissions + an Agent Access permission set

Post-setup — Human escalation   (only after item 3 is active)
  5. Employee Agent escalation           (configure canEscalate, outbound routing, a staffed General IT queue, and failure-threshold directives)

Stage 1 (Agentforce Studio enablement) is the foundation — it turns on the org-level Agentforce and Einstein GenAI toggles every agent (and the specialized employee templates) is built on, so enable it first; any agent install/activate attempted before it will fail. The three Stage 2 items are independent siblings, orderable freely once Stage 1 is done, and Specialized Agents for Employee sits right after the IT Service Employee Agent because both build employee agents from the same child skill (the broad default vs. a specialized template the user picks). Employee Agent escalation (a post-setup action on a live Employee agent) and Stage 3 (Runtime access) (automatic after any Stage 2 agent goes live — not a menu pick) are governed by their rules above.

---

Rules

• ALWAYS show "(via service-itsm-agentic-setup-agentforce-coordinate)" in the setup header
• ALWAYS present the feature menu before doing anything — do not assume which feature the user wants
• ALWAYS present the feature menu as a multi-select — accept a set of one or more features in a single interaction
• NEVER set up a feature without the user selecting it. (Explicit selection ensures the user confirms intent and avoids partial configurations if they cancel mid-flow; use the sequential-confirmation loop in the "set up everything" rule for bulk requests.)
• NEVER show features that do not have a working child skill
• If the user says "set up everything" or "all", walk through each available feature sequentially in the recommended order, confirming between each step
• Track progress across the conversation — do not re-present completed features as "Not done"
• Specialized Agents for Employee (#4) is re-selectable — each run builds a different specialized employee agent from a template the user picks. Mark the agent just built as Done, but keep #4 available to run again for additional specialized agents; do not treat a completed #4 as permanently finished the way the one-and-done items are. If the user picks #4 again, ask which specialized template to use next
• Employee Agent escalation (#5) is a post-setup action that requires the IT Service Employee Agent (#3) to be Active first — never delegate #5 before #3 has succeeded in this run (or was already done)
• NEVER advance to the next feature in the queue if the current one failed or only partially succeeded — stop and surface the failure in plain language instead
• If Agentforce Studio enablement reports the org lacks the Agentforce license (accessCheck), STOP the whole flow — this is a license/edition requirement no API can grant, and neither the Fulfiller nor the Employee Agent can succeed without it
• ALWAYS confirm the target org before delegating to any child skill, and state that the org will be modified
• Do NOT expose internal technical jargon in user-facing output. This includes Salesforce record IDs and org IDs, raw HTTP status codes (403, 500, …), API error codes (FUNCTIONALITY_NOT_ENABLED, DUPLICATE_VALUE, …), internal endpoint/API names, developer names (feature apiNames like sales-cloud-agent-studio), and CLI/tooling internals. Translate everything to plain, human-readable language. Child-skill names shown as next-step pointers are fine.
• NEVER name a feature by a bare item number in user-facing prose. The #1–#5 labels (and the menu's # column) are internal shorthand for this skill's ordering/delegation rules and a menu selection handle only (where the user replies 1, 2) — not feature names. In every message the user reads — opt-out/skip, prerequisite/blocked, progress, next-step, and the completion summary — name each feature in full, never a bare number or range like #4 or #1–#3. This matters most when a Stage 1 toggle is skipped: name the blocked and still-available features, don't say "#4 is blocked, #1–#3 can proceed". See Naming features in prose in examples/output-templates.md.
• If the user asks about Agentforce features that are not yet available (e.g., Requester agent, custom topic packs, agent metrics dashboards), tell them those features are not yet available in this orchestrator and will be added as their child skills merge

---

Verification checklist

Before emitting any menu or summary in this skill, mentally confirm each of the following. If any box is unchecked, adjust the output before sending.

• The header line ends with (via service-itsm-agentic-setup-agentforce-coordinate)
• The target org was confirmed with the user, and they were told it will be modified, before any child skill ran
• The current feature's child-skill result was verified as a full success before advancing to the next queued feature — a failed or partial result stopped the queue instead
• Only features with a working child skill are shown; placeholder features are hidden
• The feature menu is presented as a multi-select (single-select only if the user has already named a specific feature)
• Each feature row's Status column reflects the actual tracked state from the conversation (Not done, In progress, or Done) — not a hard-coded default
• Each feature row's Description matches the Salesforce Setup → Agentforce for IT Service page wording (Fulfiller and Employee agent-template descriptions verbatim, not paraphrased), the three Stage 2 rows each end with a short Setup: install line, and the AskUserQuestion option descriptions carry the same text
• The Specialized Agents for Employee row is present as a Stage 2 item shown right after the IT Service Employee Agent, and #4 is never delegated to service-itsm-agentic-setup-employee-agent-configure without first asking the user which specialized template they want (handed no name, the child skill defaults to the broad agent and duplicates #3), and it carries its Setup: install line
• The Employee Agent escalation row is present as a post-setup item shown after the Stage 2 agents, and #5 is never delegated to service-agentforce-human-escalation-configure before the IT Service Employee Agent (#3) is Active
• For a completion summary, the header line and closing line are chosen by the rubric in examples/output-templates.md (all Done → Complete; any Not done/In progress → Finished)
• A feature is being configured only because the user explicitly selected it (or is being walked through sequentially with confirmation under an "all" / "everything" request)
• Studio enablement is verified done before delegating to any Stage 2 agent (Fulfiller, Employee, or Specialized Agents for Employee)
• If any Stage 2 agent went live, Stage 3 runtime access was appended once and delegated to service-itsm-agentic-setup-agent-runtime-access-assign (skipped otherwise)
• The next action delegates to a child skill, never configures a feature inline
• No Salesforce record IDs appear in the output — human-readable names only
• No feature is named by a bare item number or range (#4, #1–#3) in user-facing prose — every feature is named in full; digits appear only as menu selection handles

---

Reference File Index

File	When to read
examples/output-templates.md	Behavior steps 3, 5, and 6 — feature menu (two-stage, multi-select), post-feature progress, and completion summary text blocks
scripts/verify-child-verdict.mjs	Behavior step 5 — run via Bash (node) to check a child skill's verdict deterministically before advancing the queue