本文へ移動
cccskills
無料GitHub で公開

territory-design-requirements

Requirements for a Salesforce Enterprise Territory Management (ETM) territory design: alignment criteria, coverage model selection, assignment rule logic, hierarchy depth and breadth, and user-to-territory ratios. Trigger keywords: territory design, territory alignment, territory model requirements, sales coverage model, territory criteria, geographic territory, named account territory, overlay territory. NOT for ETM setup steps — use admin/enterprise-territory-management. NOT for loading territory assignment data — use data/territory-data-alignment. Trigger keywords: territory design questionnaire, territory requirements document, assignment rule matrix, territory type priority, realignment plan, territory access level decision, accountAccessLevel, opportunityAccessLevel, ObjectTerritory2Association, UserTerritory2Association, Planning state model, territory acceptance tests.

インストール方法を見る

含まれるファイル(9)

  • SKILL.md28.9 KB
  • references/examples.md7.0 KB
  • references/gotchas.md15.9 KB
  • references/llm-anti-patterns.md14.6 KB
  • references/well-architected.md9.3 KB
  • references/worked-examples.md25.6 KB
  • scripts/check_territory_design_requirements.py30.4 KB
  • templates/territory-design-requirements-template.md12.8 KB
  • templates/territory-design.yaml4.9 KB

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。

Territory Design Requirements

This skill activates when a practitioner needs to gather, evaluate, or document the requirements for a Salesforce Enterprise Territory Management (ETM) territory design before configuration begins. It covers alignment criteria selection, coverage model type, assignment rule logic, geographic considerations, hierarchy depth and breadth, per-object access-level decisions, realignment planning, and user-to-territory ratio targets. Use this skill before invoking admin/enterprise-territory-management for configuration.

The deliverable is a design that survives contact with the metadata. Every decision below has a field it lands in, and that field is named.


Before Starting

Gather this context before working on anything in this domain:

  • ETM is the target platform. ETM is not Legacy Territory Management. The metadata types the design must land in are Territory2Model, Territory2Type, Territory2, Territory2Rule and Territory2Settings (Metadata API Developer Guide). A design expressed only in slide boxes cannot be checked against them.
  • The most common wrong assumption is that territory hierarchy mirrors role hierarchy. ETM hierarchy is independent of role hierarchy. Territory hierarchy drives the Territory and TerritoryAndSubordinates sharing groups and territory forecast rollups; role hierarchy drives manager visibility in the standard pipeline. Conflating them produces a structure that has to be re-cut every time the org chart moves.
  • A territory assignment rule holds at most 10 rule items. "A territory rule can have up to 10 rule items" (Metadata API Developer Guide, Territory2Rule › Usage). This is a hard ceiling on the rule, not a soft performance guideline — a segmentation that needs an eleventh predicate needs a proxy field on Account instead.
  • Territory type priority: the highest integer wins, and it must be unique. "The account-assigned territory whose territory type priority is highest is then assigned to the opportunity. The priority field value on each territory type must be unique." (Territory2Type › priority). A tie assigns no territory to the opportunity.
  • Territory access is always additive. Access levels grant users access to records "that are assigned to this territory and are otherwise inaccessible" (Territory2 › accountAccessLevel). ETM cannot restrict below the org-wide default — a design that relies on territories to limit access belongs in admin/sharing-and-visibility.
  • Territory count ceilings: UNVERIFIED (2026-09-05). The commonly cited "1,000 territories per model, higher on request" figure does not appear in the Salesforce App Limits Cheat Sheet (a case-insensitive grep for "territor" across it returns zero lines) nor in the Territory2* sections of the Metadata API guide or Object Reference. Confirm the current allocation with Salesforce before designing to a ceiling; treat any number you carry into the requirements document as an assumption with an owner.

Questions to Ask Before Configuring

Ask these before writing a single territory name. Each maps to a documented platform behaviour that otherwise surfaces after the model is Active and the recalculation is already running.

AskWhy it mattersWhat a good answer adds
"Which dimensions actually decide who covers an account — and which are just how you report on it?"Only a dimension that is a field on Account can become a Territory2Rule rule item; a reporting dimension does not need a territoryA short list of real Account fields, separated from the dimensions that belong in a report grouping
"For each dimension, name the Account field, its API name, and how populated it is today."Territory2RuleItem.field operates on a standard or custom object field; a null field silently matches nothing and the account lands nowhereField API names plus a populated-row count, which sizes the catch-all territory before it is needed
"Can one account be covered by two teams at once, and if so is that deliberate?"An account matching two rules is assigned to both territories, and two territories of the same type on one account assign no territory to the opportunity at allAn explicit overlay decision plus a unique priority integer per type
"What are the org-wide defaults for Account, Opportunity, Contact and Case today?"OWD narrows which access-level values are legal — with Public Read/Write on accounts the only valid accountAccessLevel values are Edit and All, and for some objects the element must be omittedA per-object access decision written in Metadata API spelling, not Setup labels
"Does the forecast roll up by territory, and who reads it at each level?"Territory forecast rollup follows the hierarchy of the active model, so a hierarchy level with no reader is a level that exists only to be maintainedOne named forecast consumer per hierarchy level, or that level deleted
"How often does this alignment change, and what is the cutover window?"A model is created in Planning, only Planning or Active models can be deployed, and archiving is one-way — realignment is a build-alongside-then-cut-over exercise, not an editA realignment cadence, a named runner for the rule run, and a fiscal date the cutover must precede
"Who is allowed to change a territory after go-live, and at which node?"TerritoryAdminAssignment (API 63.0+) delegates hierarchy, membership and record-association rights per territory subtree to a user holding Administer Territory OperationsA delegation table, instead of every change queueing behind one system administrator

What a proper configuration adds over just doing it: the design arrives at build time as named Account fields, unique type priorities, legal access-level values and a dated cutover plan, so the ETM build is a transcription rather than a second round of discovery — and the acceptance tests are already written when the rule run finishes.


Core Concepts

Coverage Model Types

ETM supports four coverage patterns, and most orgs run a combination:

Geographic: Territories defined by location criteria — BillingState, BillingCountry, BillingPostalCode, or a custom region picklist. Best for field sales organisations with clear geographic accountability. The Territory2RuleItem operation enum also carries within, documented as "DISTANCE criteria only", so radius-style criteria exist where a geolocation field does.

Named Account: Specific accounts assigned to specific reps regardless of location. Implemented via an Account field (a named-account flag is cheaper than an Account Name match list) or by manual assignment, which lands as ObjectTerritory2Association.AssociationCause = Territory2Manual rather than Territory2AssignmentRule.

Industry Overlay: Cross-functional coverage granting additional visibility across other territories by industry, product line, or segment. Overlays stack on primary coverage — members gain additive access.

Hybrid: Geographic primary coverage plus named account or overlay territories. Most enterprise orgs end up here. Requires an explicit, unique priority integer per territory type, because that integer is what decides opportunity territory assignment.

Hierarchy Design Principles

The territory hierarchy sets parent-child relationships that control two things: how forecast data rolls up, and how the TerritoryAndSubordinates group is built. That group is a public group of "users in a specified territory, and users in territories subordinate to the specified territory" (Metadata API Developer Guide, FolderShare › sharedTo; the same pairing appears on territory sharing rules), and it is what propagates record access up a branch.

Depth: UNVERIFIED (2026-09-05): the widely repeated "3–5 levels" recommendation is not stated in the Metadata API guide, Object Reference or App Limits Cheat Sheet. What is checkable is the practice test behind it — every level must have a named person who reads the forecast at that node. Levels that fail that test are maintenance with no reader.

Breadth: Span of control per level should reflect the real management structure. A node with 50 direct children is not auditable and usually means a missing intermediate level.

Independence from role hierarchy: Territory2.ParentTerritory2Id is the only parenting mechanism ETM has, and it has no relationship to UserRole. Design the two hierarchies separately against their own purposes even where they end up similar in shape.

Assignment Rule Design

Account assignment rules are filter-based criteria evaluated against Account field values. A rule belongs to the model, and its link to a territory is a separate ruleAssociations entry on each Territory2 carrying an inherited flag — "Rule inheritance flows from the parent territory where the rule is created to the rule's descendent territories… A local rule is created within a single territory and affects that territory only." The design must therefore state, per rule, both the territories it serves and whether it is inherited.

Design constraintGrounded in
Max 10 rule items per ruleMetadata API, Territory2Rule › Usage: "A territory rule can have up to 10 rule items."
Operations available: equals, notEqual, lessThan, greaterThan, lessOrEqual, greaterOrEqual, contains, notContain, startsWith, includes, excludes, within (DISTANCE only)Territory2RuleItem › operation
Combine items with booleanFilter, e.g. (1 AND 2) OR 3; "Numbering must start at 1 and must be contiguous"Territory2Rule › booleanFilter
Rule item order is positional — "The sort order of rule items is implicitly derived from the position of the rule items in the XML"Territory2Rule › Usage
objectType is Account (Lead assignment is a separate Territory2Settings.supportedObjects switch)Territory2Rule › objectType; Territory2SupportedObject › objectType
An account matching several rules lands in several territoriesObjectTerritory2Association, one row per assignment
Blanket coverage: something must own the accounts no rule matches, or they appear in no territory forecastObjectTerritory2Association exists only where an assignment happened

Date-field criteria: UNVERIFIED (2026-09-05) — the long-standing practitioner claim that ETM assignment rules reject Date and DateTime fields is not stated in the Metadata API guide (Territory2RuleItem.field is documented only as "The standard or custom object field that the rule item operates on") or the Object Reference. The operation enum does carry the comparison operators a date would need. Treat a date-based criterion as unproven until you have configured one in a sandbox, and prefer a quarter/month proxy picklist regardless, because a date predicate goes stale the moment the date passes while a rule run happens on someone's schedule.

Access-Level Decisions

Access levels are a design decision, not a build detail: they are per territory and per object, and OWD constrains which values are legal.

ElementValid valuesRule from the guide
accountAccessLevelRead, Edit, AllWith a Public Read/Write account sharing model, only Edit and All are valid
opportunityAccessLevelNone, Read, EditSpecify no value if the case/opportunity sharing model is Public Read/Write
caseAccessLevelNone, Read, EditSame omission rule as opportunity
contactAccessLevelNone, Read, EditSpecify no value if the contact model is Public Read/Write or Controlled By Parent
objectAccessLevels (API 57.0+)accessLevel of Read, Edit, Transfer, All plus an objectTypeThe org-level counterpart, Territory2SupportedObject, documents Lead as the only supported object type

Omitting an element falls back to the matching default*AccessLevel in Territory2Settings. The Metadata API spellings above are not the labels the same values carry as SOQL picklists on the Territory2 object (Read Only / Read/Write / Owner for accounts; Private / Read Only / Read/Write elsewhere) — the requirements document should carry the Metadata API spelling, because that is what deploys.

User-to-Territory Ratio

UNVERIFIED (2026-09-05): the "approximately 3:1 territories per user" target is a field heuristic, not a documented Salesforce figure — it is absent from the App Limits Cheat Sheet, the Metadata API guide and the Object Reference. Use it as a review trigger rather than a rule:

  • Ratios far above 3:1 mean many leaf territories per person: more rules to maintain, more ObjectTerritory2Association rows to reconcile, fragmented forecast nodes.
  • Ratios below 1:1 mean coverage accountability is ambiguous — several people in one territory with no documented split.
  • A user may belong to several territories; count each UserTerritory2Association row, and remember IsActive on that row can be false while the row still exists.

Test the ratio against account volume per territory, not against headcount alone.


Common Patterns

Geographic Primary Coverage Model

When to use: Clear geographic accountability, field reps own a region, and assignment can be determined from BillingState or BillingCountry.

How it works:

  1. Define territory types: National → Regional → Sub-Regional, each with a unique priority integer.
  2. Build the hierarchy with parentTerritory set to the parent's developer name — "use the developer name. Do not use the 'fully qualified' name" (Territory2 › parentTerritory).
  3. Create mutually exclusive rules at the leaf level: BillingState equals CA, BillingState equals OR.
  4. Parent territories need no rules of their own — access flows through the TerritoryAndSubordinates group; an inherited rule is a separate, deliberate choice recorded in ruleAssociations.inherited.
  5. Define a catch-all territory for accounts matching no leaf rule, including blank billing addresses.

Why not a flat structure: A single level of 50 nodes gives forecast rollup nothing to aggregate on, and every manager view becomes a custom report.

Named Account Overlay Model

When to use: A subset of accounts belongs to strategic reps regardless of location, and those accounts are also inside geographic territories.

How it works:

  1. Create a distinct territory type with a higher priority integer than the geographic type — the highest integer wins opportunity territory assignment.
  2. Create one territory per named-account rep or book.
  3. Assign accounts by a flag field on Account (one rule item, stable) or manually; manual lands as AssociationCause = Territory2Manual and AccountShare.RowCause = Territory2AssociationManual, rule-driven lands as Territory2AssignmentRule and RowCause = Territory. Those two values are how the audit tells them apart.
  4. Both reps hold access — territory access is additive and does not change account ownership.
  5. Keep the two types' priorities unique. Two territories of the same type on one account assign no territory to the opportunity.

Why not a sharing rule: A sharing rule grants access without producing an ObjectTerritory2Association, so the account never reaches territory forecasting. Where territory forecasting is in scope, membership is the mechanism.


Decision Guidance

SituationRecommended ApproachReason
Reps cover fixed geographic regions with no overlapGeographic model with mutually exclusive BillingState/BillingCountry rulesCriteria are stable, auditable, and one rule item each
Enterprise reps own accounts across geographiesNamed-account territory type overlaid on geographic primary coverage, with the higher priority integerBoth coverage layers keep access; the higher integer decides the opportunity
Industry specialists cover accounts across all regionsOverlay territory type per vertical, unique priorityAccess is additive; primary and overlay coexist
Segmentation needs an 11th predicate on one territoryCollapse to a proxy field on Account (e.g. Territory_Segment__c picklist)A rule holds at most 10 rule items (Metadata API, Territory2Rule › Usage)
Criteria combine AND and OROne rule with a booleanFilter such as (1 AND 2) OR 3Numbering must start at 1 and be contiguous, and item order is positional
Assignment criteria are date-basedUse a quarter/month proxy picklist maintained by automationA date predicate is stale between rule runs; the field-type restriction itself is UNVERIFIED (2026-09-05)
Named account list changes monthly or fasterFlag field with one rule item, or manual assignmentAvoids editing rules on every list change; manual assignments are identifiable by AssociationCause
Two territories of the same type can hold one accountSplit into two types with unique priorities, or make the criteria exclusiveA same-type tie assigns no territory to the opportunity (Territory2Type › priority)
Alignment changes every fiscal yearPlan a Planning-state model built alongside the active one, with a dated cutoverArchiving is one-way and only Planning or Active models can be deployed
Requirement reads "reps should see only their accounts"Return it to admin/sharing-and-visibility as an OWD requirementTerritory access grants access to records "otherwise inaccessible" — it never removes any

Recommended Workflow

  1. Run the questionnaire in references/worked-examples.md §1 with sales leadership and sales ops in the room. Record answers verbatim; a dimension nobody can name an Account field for is not yet a requirement.
  2. Fill templates/territory-design.yaml — territory rows with type, priority, parent, rules or an explicit manual_assignment: true, per-object access levels, and the realignment plan. templates/territory-design-requirements-template.md is the prose companion for the parts a linter cannot judge.
  3. Draft the hierarchy and the assignment-rule matrix (references/worked-examples.md §2–§4). For each rule: field API name, operation from the documented enum, value, the territories it serves, and whether it is inherited. Count the rule items — the ceiling is 10 per rule.
  4. Decide the access levels per object per territory against the org's current OWD, using the table in Core Concepts and the enum in references/worked-examples.md §5. Write the Metadata API spelling.
  5. Write the realignment runbook (references/worked-examples.md §6): which model states the plan passes through, who runs the rules, and the fiscal date the cutover must precede.
  6. Lint the design: python3 scripts/check_territory_design_requirements.py --manifest-dir <design-dir>. It fails on missing type/priority/parent, duplicate priorities, territories with neither a rule nor a manual-assignment flag, access levels outside the documented enums, two rules with identical criteria in one territory, a missing realignment plan, and any Territory2Rule XML in the folder whose ruleItems use an operation outside the enum.
  7. Attach the acceptance tests (references/worked-examples.md §7) to the handoff and pass the package to admin/enterprise-territory-management. The tests are SOQL over ObjectTerritory2Association and UserTerritory2Association; they are the definition of "the build matches the design".

Review Checklist

Run through these before marking requirements complete:

  • Coverage motion is confirmed (geographic, named account, overlay, or hybrid)
  • Every alignment dimension maps to a named Account field API name with a known population rate
  • Every hierarchy level has a named forecast consumer; levels without one are removed
  • Territory hierarchy is documented independently of the role hierarchy
  • Territory types are named and each priority integer is unique — and the highest one is the type intended to win opportunity assignment
  • No two territories of the same type can hold the same account (a tie assigns no territory)
  • No rule exceeds 10 rule items; every operation is in the documented enum; booleanFilter numbering starts at 1 and is contiguous
  • Every territory has at least one assignment rule or an explicit manual-assignment decision with an owner
  • ruleAssociations.inherited is decided per rule, not left to the builder
  • Access levels are chosen per object per territory and are legal for the current OWD, in Metadata API spelling
  • Catch-all coverage exists for accounts matching no rule, including blank billing addresses
  • A realignment plan names the model states, the rule runner, and the cutover date
  • Acceptance tests over ObjectTerritory2Association / UserTerritory2Association are written before build starts
  • Any territory-count ceiling carried in the document is marked as an assumption with an owner (see Before Starting)
  • python3 scripts/check_territory_design_requirements.py --manifest-dir <dir> reports no ERROR findings

Salesforce-Specific Gotchas

Non-obvious platform behaviors that cause real production problems. Full form in references/gotchas.md.

  1. The priority integer runs the opposite way to most people's intuition — the highest integer wins opportunity territory assignment, values must be unique per type, and a same-type tie assigns nothing at all.
  2. Ten rule items is a hard ceiling on the rule — the eleventh predicate needs a proxy field on Account, not a second attempt at the rule.
  3. Overlapping criteria assign an account to every matching territory — deliberate for overlays, almost never deliberate for primary coverage.
  4. A rule change assigns nothing on its own — rules run automatically on record create and edit; a change to the rule itself needs a user-initiated run, and Metadata API cannot start one.
  5. Two flags silently stop rules from evaluating a record — IsExcludedFromRealign on the record and tm2BypassRealignAccInsert in the org settings.
  6. Territory access can only add — it grants access to records that are "otherwise inaccessible"; restriction is an OWD conversation.
  7. Archiving is one-way and a same-named archived model blocks the deploy — realignment is build-alongside-then-cut-over.

Output Artifacts

ArtifactKindDescription
Territory design requirements documentMarkdownCoverage model, hierarchy shape, criteria, ratio analysis, open decisions (templates/territory-design-requirements-template.md)
territory-design.yamlYAMLThe machine-lintable design: territories, types, priorities, parents, rules, access levels, realignment plan (templates/territory-design.yaml)
Assignment rule matrixTablePer rule: field, operation, value, territories served, inherited flag, rule-item count
Territory hierarchy tableTableLevels, parent-child relationships, territory type, named forecast consumer per level
Access-level decision tableTablePer territory per object, in Metadata API spelling, checked against current OWD
Realignment runbookMarkdownModel states, rule runner, cutover date, rollback position
Acceptance testsSOQLQueries over ObjectTerritory2Association, UserTerritory2Association, Territory2AlignmentLog that prove the build matches the design

Reference Files

FileRead it when
references/worked-examples.mdFilling in the questionnaire, the hierarchy table, the rule matrix, the access decisions, the realignment runbook or the acceptance tests — one B2B SaaS scenario carried end to end
references/gotchas.mdA design looks reasonable on paper and you want to know how it fails after activation
references/examples.mdShaping a geographic model or a named-account overlay, and sizing the ratio
references/well-architected.mdTrading off depth, territory count and access breadth — and for the source list
references/llm-anti-patterns.mdReviewing generated territory-design advice, especially anything asserting a priority direction or a field-type restriction
templates/territory-design.yamlStarting the design artefact the checker lints
templates/territory-design-requirements-template.mdWriting the prose requirements document around it

Related Skills

  • admin/enterprise-territory-management — the build: Territory2* metadata, the deploy path, activation, rule runs and verification SOQL. Read its references/metadata-examples.md rather than duplicating XML here
  • admin/sharing-and-visibility — OWD, role hierarchy and sharing rules set the floor that territory access adds to; any "restrict to their own accounts" requirement lands here
  • admin/opportunity-management — opportunity stages, forecast categories and the pipeline the territory forecast rolls up
  • admin/requirements-gathering-for-sf — the general discovery practice this skill specialises; its requirements catalogue is the parent artefact a territory design plugs into
  • admin/collaborative-forecasts — forecast types and quotas, once the territory forecast type is in scope
  • admin/role-hierarchy-design — the structure the territory hierarchy is deliberately independent of
  • data/territory-data-alignment — bulk-loading UserTerritory2Association and realignment data at volume

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