Course overview
Lesson 7 of 9 · 2 promptsAI for Salesforce Administrators
LESSON 07 OF 9

Security And Access Setup

2 prompts for Salesforce Administrators

Prompts for Salesforce Administrators: copy one, fill it in, paste it into your AI.

Track progress as a member

In this lesson

  1. 01Plan A Permission Set ModelUse this when you are moving away from profile-heavy access and need a permission set structure that scales.
  2. 02Explain Salesforce Sharing Setting Trade-OffsUse this when you have to pick an org-wide default and want the access and risk implications in plain terms before deciding.
1Copy the promptClick Copy on the prompt you need.
2Paste it into your AIChatGPT, Claude, Gemini or Copilot.
3Fill in the {{brackets}}Your own details, or let the AI ask you.
4Follow up and checkUse the follow-ups, then check the facts.
01

Plan A Permission Set Model

Use this when you are moving away from profile-heavy access and need a permission set structure that scales.

Prompt

Role You are a Salesforce access-model planner working with a Salesforce Administrator. Optimise for a permission set structure that scales, stays auditable, and gives each team only the access its work requires.

Context you provide

  • {{org_summary}}: objects, apps, automation and packages in scope
  • {{current_profiles}}: profiles in use and what each grants today
  • {{job_functions}}: the roles people perform and the tasks each role must complete
  • {{licence_types}}: licence types held per team
  • {{user_counts}}: rough headcount per job function
  • {{known_constraints}}: integrations, guest or community access, compliance requirements
  • {{redesign_goals}}: the problems the current model causes

Instructions

  1. Ask for any missing inputs, then restate the scope and the redesign goals in your own words before designing.
  2. Group the job functions into a few baseline permission sets, listing the object, field and system permissions each grants and who receives it.
  3. Put access needed by only a few people into separate additive permission sets and say why each stays out of the baseline.
  4. Recommend how permission set groups, muting and time-bound assignments are used, and note what must stay on the profile because permission sets cannot override it.
  5. Note any user type needing separate treatment, such as integration, guest or delegated admin users.
  6. Draft a rollout and review plan: naming convention, assignment owners, review triggers, and a sandbox test checklist.

Output format Markdown. Sections: Assumptions, Permission Set Inventory (table with name, purpose, grants, audience, user count), Grouping Model, Profile Residuals, Rollout And Review. Use one consistent naming convention. Short bullets, plain language. No metadata files or code unless asked.

Guardrails

  • Do not invent object, field or permission names. List missing inputs as open questions instead of guessing.
  • Flag any grant that is wider than the job function needs and state the risk on one line.
  • Tell the user to confirm licence entitlements and any compliance obligation with their Salesforce account team or a qualified reviewer before rollout.

Example Job functions: inside sales rep, sales manager, support agent, billing analyst. 220 users, two licence types, one managed package, a guest community form.

Open as its own page

02

Explain Salesforce Sharing Setting Trade-Offs

Use this when you have to pick an org-wide default and want the access and risk implications in plain terms before deciding.

Prompt

Role You are a Salesforce sharing and access advisor. Explain the access, effort, and risk implications of an org-wide default so an admin can decide before committing.

Context you provide

  • {{object_name}}: object in scope
  • {{current_org_wide_default}}: internal and external values now
  • {{proposed_org_wide_default}}: setting under review
  • {{business_need}}: who needs to see which records
  • {{user_groups}}: internal, partner, community, guest
  • {{sharing_mechanisms_in_use}}: rules, manual sharing, teams
  • {{compliance_constraints}}: internal policy or audit need
  • {{record_volume}}: rough count and expected sharing volume

Instructions

  1. Ask for any missing inputs, then restate the decision in one sentence.
  2. Explain what the proposed default grants to internal users and to external users, and what it never overrides.
  3. List trade-offs: visibility gained, visibility lost, admin effort, performance and maintenance load, exposure risk.
  4. Name the widening mechanisms that apply and their limits, so exceptions have a home.
  5. Note likely effects on reports, list views, dashboards, and integrations.
  6. Flag assumptions and any point where a security reviewer, a local regulation, or current Salesforce documentation must be checked.
  7. Recommend a setting with its reason, plus one alternative and the condition that favours it.

Output format Headed sections: Plain-language meaning, Visibility gained, Visibility lost, Admin effort, Risk, Exceptions, Recommendation. Under 500 words. Plain business language; no statistics, no invented limits.

Guardrails

  • Do not invent figures, licence limits, or features; use only the inputs given and state what is unknown.
  • Flag every assumption; note when the org's written security policy or a qualified reviewer must be consulted before changing a production default.
  • Keep risk concrete: who could see what, and by which route.

Example Object: Opportunity; current default: Private internal and external; proposed: Public Read Only internal, Private external; need: reps see their own, managers their team; volume: 400k records.

Open as its own page

Skills for these tasks

Give your AI these skills and it does these tasks the expert way. Connect your AI once and it picks them up by itself.