thinkhard.SI

Prompt repository Hiring & people

Write a social media policy for staff

Employees post about work on their own accounts, and you need a clear policy built from your decisions that a lawyer can check before it goes out.

Registry no.
PPL-039
Collection
Hiring & people
Laid out for
Any chat model
Score
10 out of 10
Blanks to fill
6
Added

Hiring & peopleAny chat

The prompt

Score 10 of 10, Excellent

Use it when Employees post about work on their own accounts, and you need a clear policy built from your decisions that a lawyer can check before it goes out.

  • {{company}}
  • {{reviewer}}
  • {{decisions}}
  • {{related_policies}}
  • {{locations}}
  • {{official_voices}}
Full prompt and why it works
Prompt
Write a social media policy for the employees of {{company}}, built from the decisions below, so that staff know what is up to them on their own accounts, who may speak for the company, and what happens if a post crosses a line. Employees will read it once when it comes out and look it up again when they are unsure about a post, so each rule has to be clear enough to apply to a real post without asking HR. {{reviewer}} will review it before it is published.

Our decisions on social media:
"""
{{decisions}}
"""

Policies this one should point to rather than repeat (conduct, anti-harassment, confidentiality, discipline):
"""
{{related_policies}}
"""

Where our employees work: {{locations}}
Who runs our official accounts and who may speak to the press: {{official_voices}}

Use these headings, in this order:
1. Who this covers and why we have it, in two sentences.
2. Your own accounts: what is up to you, and the few things that are not.
3. Speaking for the company: who may, and what to do if a journalist, customer or follower asks you something.
4. Saying you work here: when to mention it, for example when you post about our products or a competitor's.
5. Confidential information: what counts, taken from our decisions, with two examples of each kind.
6. Respect online: how our conduct and anti-harassment policies apply to posts and messages about colleagues, customers and partners.
7. Social media during work hours and on work devices.
8. Questions, reporting a problem, and what happens if the policy is broken (point to the discipline policy).

Keep it to 600 to 900 words, written to the employee as "you", in plain language. A good rule names a behavior, not an attitude: "Do not post photos of customer documents or screens" rather than "Use good judgment."

Employment law in many places protects employees who talk with each other about pay, hours and working conditions, including on social media; in the United States this falls under the National Labor Relations Act. So do not write rules that ban criticizing the company, discussing pay or benefits, or talking about working conditions, and do not require approval before employees post about their jobs. Where one of our decisions might restrict that kind of speech, keep it out of the policy text and list it for {{reviewer}} instead. Do not invent rules, past incidents or consequences that our decisions do not include, and do not state what the law requires in any place.

If our decisions do not cover a heading, keep the heading and write [DECISION NEEDED: what to decide] under it. After the policy, add "For legal review": each rule that may be limited by the law in {{locations}}, with one line on why, and every [DECISION NEEDED] marker.

Variables

{{company}}
Your company or organization name.
{{reviewer}}
Who reviews the policy before it goes out, for example "our employment counsel".
{{decisions}}
What leadership has decided: what counts as confidential, who may speak for the company, rules for official accounts, anything employees must not post.
{{related_policies}}
Names or short summaries of the related policies to point to, such as conduct, anti-harassment and discipline.
{{locations}}
The countries or states where your employees work.
{{official_voices}}
Who runs official accounts and who may talk to the press, for example "the marketing team; media questions go to the head of communications".

Why it works

  • What to avoid. Rules that would ban criticizing the company or discussing pay and working conditions stay out of the text and go to the reviewer, because many places protect that kind of speech.
  • What good looks like. Each rule has to name a behavior an employee can recognize in a real post, not an attitude such as good judgment.
  • If unsure. Headings your decisions do not cover get a [DECISION NEEDED] marker, so the policy never carries a rule nobody chose.

How to use it

  1. Press Fill in blanks and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/staff-social-media-policy

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "staff-social-media-policy"}, or just ask for it by name.