本文へ移動
cccskills
無料GitHub で公開

flow-orchestration-admin

Flow Orchestration admin: stage configuration, step assignment, background steps, interactive steps, evaluation flows, work items, pause/resume. NOT for approval routing a standard approval process would handle — use admin/approval-processes.

インストール方法を見る

含まれるファイル(7)

  • SKILL.md6.6 KB
  • references/examples.md6.6 KB
  • references/gotchas.md8.9 KB
  • references/llm-anti-patterns.md4.5 KB
  • references/well-architected.md5.9 KB
  • scripts/check_flow_orchestration_admin.py4.0 KB
  • templates/flow-orchestration-admin-template.md635 B

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Flow Orchestration Admin

Activate when building multi-step, multi-actor processes with Flow Orchestration in Setup: pause-and-wait workflows that cross users, queues, and background systems. Flow Orchestration is the successor to Approval Processes for complex branching; it is a no-code first-class orchestrator with stages, steps, and work items.

Before Starting

  • Confirm the process actually needs orchestration. Single-user approvals → Approval Process is fine. Multi-actor with parallel work, waits, and conditional branching → Orchestration.
  • Map the actors and handoffs. Every hand-off (user A → user B, system → user, user → system) is a Step. Diagram them before opening Flow Builder.
  • Decide on interactive vs background steps. Interactive step = a user opens a work item screen. Background step = a Flow runs headless. Wrong choice produces confused users or stuck work.

Core Concepts

Stage → Step → Work Item

An orchestration consists of Stages (run sequentially). Each Stage has Steps (can run in parallel). Each Step is either Interactive (produces a Work Item a user picks up) or Background (runs a flow). Stages advance when all steps complete — including conditional skips.

Evaluation Flow

A Stage's entry and exit conditions are evaluated by an Evaluation Flow — a small autolaunched flow returning a boolean. Used to decide whether to skip a stage, loop back, or proceed.

Work Items

Interactive steps produce WorkItem records assigned to a user or queue. The assignee opens the Work Items list, picks an item, completes the associated screen flow, and the step advances.

Pause and resume

Orchestration pauses at Interactive Steps waiting for user action and at wait elements inside Background Steps. Resumption is automatic on work item completion or wait-condition satisfaction.

Common Patterns

Pattern: Parallel review with consolidated decision

Stage 1: Background — populate context. Stage 2: Three Interactive Steps in parallel (legal, security, finance review), each producing a Work Item. Stage 3: Evaluation Flow consolidates approvals. Stage 4: Background — enact the decision.

Pattern: Long-wait with external callback

Background Step kicks off an external process via callout and stores a correlation ID. Orchestration waits on a Platform Event the external system publishes on completion. The event resumes the step.

Pattern: Conditional rework loop

Stage 4 Evaluation Flow detects a rejection. The orchestration loops back to an earlier Stage to rework. Use conditional navigation (branches between stages) carefully — infinite loops are possible.

Decision Guidance

SituationRecommended ApproachReason
Single approver, simpleApproval ProcessLower setup cost
Multi-actor parallel workFlow OrchestrationApproval Process cannot branch
Wait for external callbackOrchestration + Platform EventDecouples from source
Time-based timeoutStage with Evaluation Flow + waitShipped pattern
Round-robin assignmentWork item to queue + assignment rulesDon't hand-code

Recommended Workflow

  1. Whiteboard the process: actors, decision points, handoffs, SLAs.
  2. Decide Approval Process vs Orchestration with standards/decision-trees/automation-selection.md.
  3. Build required screen flows (for Interactive Steps) and autolaunched flows (for Background / Evaluation).
  4. Create the Orchestration in Flow Builder; lay out Stages and Steps; wire Evaluation Flows.
  5. Configure work item assignment (user, queue, role-based) and notifications.
  6. Test in a sandbox with full actor list; validate parallel-stage timing and evaluation branches.
  7. Deploy; monitor the Orchestrations tab; set up alerts for work items aging past SLA.

Review Checklist

  • Stage and step diagram matches whiteboarded process
  • Evaluation Flows tested for each branch
  • Work item assignment routes to correct queue/user
  • Timeouts in place for long-wait stages
  • Fault paths defined on Background Steps
  • Monitoring view available to ops team
  • Training doc for users completing Interactive Steps

Salesforce-Specific Gotchas

  1. Parallel steps finish when ALL complete. A slow reviewer holds up the whole stage; consider timeouts + reassignment.
  2. Evaluation Flows run every entry and exit. Expensive logic in an Evaluation Flow runs many times across a long orchestration; keep them lean.
  3. Deleting an in-flight Orchestration metadata breaks running instances. Always retire with a migration plan; do not delete while instances are active.

Output Artifacts

ArtifactDescription
Orchestration diagramStages, steps, branches, evaluation flows
Work item assignment specQueue, user, or role per step
Screen + autolaunched flow inventoryAll flows the orchestration invokes
Monitoring + SLA runbookOps procedure for aging work items

Related Skills

  • admin/approval-processes — simpler alternative for linear approvals
  • flow/flow-record-triggered-patterns — underlying flow building
  • admin/platform-events-for-admins — external-callback resume pattern

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Use this skill when writing test-first, behavior-driven acceptance criteria in Given/When/Then format for a Salesforce user story. Covers happy path, edge cases, negative paths, permission boundaries, and data-state preconditions so the AC block can drive UAT scripts and Apex test design downstream. Trigger keywords: given when then, gherkin, behavior driven AC, test first acceptance criteria, scenario outline, BDD acceptance criteria. NOT for the user-story format itself (use admin/user-story-writing-for-salesforce). NOT for UAT script writing (use admin/uat-test-case-design). NOT for Apex test method generation (use agents/test-class-generator). NOT for high-level UAT planning (use admin/uat-and-acceptance-criteria). More triggers: acceptance criteria record, ac_id, req_id on a criterion, artefact under test, negative case for a rule-type requirement, no-match fall-through criterion, oracle for a Then clause, test type apex flow manual, criteria linter, check_ac_format.py, --manifest-dir.

日本語の概要は準備中です。原文の説明を表示しています。

PranavNagrecha/AwesomeSalesforceSkills192026年10月4日 更新

Task and Event objects: polymorphic WhatId/WhoId, Activity object model, ActivityHistory vs OpenActivity, activity timeline customization, bulk task creation, Einstein Activity Capture boundaries. NOT for turning on EAC or calendar sync — use admin/einstein-activity-capture-setup. NOT for Email-to-Case — use admin/case-management-setup. Covers ActivitiesSettings metadata, enableActivities, shared Task/Event custom fields and FLS, TaskStatus value sets, Shared Activities and TaskRelation/EventRelation, recurring and child activities, and partial-success bulk DML on Task.

日本語の概要は準備中です。原文の説明を表示しています。

PranavNagrecha/AwesomeSalesforceSkills192026年10月4日 更新

Design Invocable Apex actions that return deterministic, agent-friendly errors instead of surfacing raw exceptions to the LLM. NOT for authoring the action itself — @InvocableMethod schema, labels, security context — use agentforce/custom-agent-actions-apex. NOT for the Apex tests that force each error branch — use agentforce/agent-action-unit-tests.

日本語の概要は準備中です。原文の説明を表示しています。

PranavNagrecha/AwesomeSalesforceSkills192026年10月4日 更新

Use when designing how an Agentforce agent extracts structured input parameters (slots) from a user's natural-language utterance to invoke an action. Triggers: 'agent extract date from user', 'agent action argument extraction', 'agentforce slot filling', 'agent invocable input mapping', 'agent fails to fill required parameter'. NOT for action authoring (use agentforce/agent-actions) or for prompt-template variable binding (use agentforce/prompt-builder-templates).

日本語の概要は準備中です。原文の説明を表示しています。

PranavNagrecha/AwesomeSalesforceSkills192026年10月4日 更新

Apex test patterns for @InvocableMethod agent actions: per-reason-code branch coverage, bulk safety, callout mocking, deterministic assertions. NOT for testing whether the agent routes to the right topic or answers well — use agentforce/agent-testing-and-evaluation. NOT for designing the error envelope the tests assert on — use agentforce/agent-action-error-handling.

日本語の概要は準備中です。原文の説明を表示しています。

PranavNagrecha/AwesomeSalesforceSkills192026年10月4日 更新

Designing or reviewing Agentforce actions: Flow actions, Apex invocable actions, prompt-template actions, action naming, input and output contracts, confirmation requirements, and safe error behavior. Triggers: 'agent actions', 'build an agent that can look up an order', 'make the agent able to check order status', 'give the agent a capability', 'agent that can retrieve a record', 'flow action for agent', 'agent invocable action', 'action schema design'. NOT for writing the Apex class itself — use agentforce/custom-agent-actions-apex. NOT for the reason_code error envelope that stops an agent retry loop — use agentforce/agent-action-error-handling.

日本語の概要は準備中です。原文の説明を表示しています。

PranavNagrecha/AwesomeSalesforceSkills192026年10月4日 更新

PranavNagrecha のスキルをすべて見る

このスキルの問題を報告する