Hr Code Of Conduct

---
name: hr-code-of-conduct
description: Writes a Code of Conduct that turns your values into observable behaviours, with a conflicts of interest and relationships section and a disclosure route, privacy at work, a speak-up route, a non-retaliation statement and what happens after a breach. Use for "run hr-code-of-conduct", "code of conduct", "workplace relationships policy", "conflict of interest disclosure", "speak up policy", "non-retaliation statement", "manager dating a report", "turn our values into behaviours", part of the AI for HR Pack by Polar Bear.
---

# Code of Conduct

## When To Use
A manager dates a report and signs off her promotion, and nothing written says what should have happened. Use this when your values sit on a wall but nobody can say which behaviour crosses a line, who to tell, or what happens to the person who speaks up.

## When Not To Use
If someone has already raised a concern, run Employee Relations Intake; a code will not handle a live case. For any other single rule (expenses, remote work, equipment), run HR Policy.

## Inputs
- Your values as written today, and where your people work (country, and state or province where it matters).
- The situations to cover (relationships, outside interests, gifts, recording at work, harassment including by clients) and your current routes for raising concerns.
If you have none of this, I start from the six sections below with [placeholders] for every value and route, and mark the output as a first draft.

## Approach
The speak-up and non-retaliation parts follow GOV.UK whistleblowing guidance for employers (gov.uk/guidance/whistleblowing-guidance-for-employers); harassment, including by third parties, follows the EHRC technical guidance (equalityhumanrights.com/guidance/sexual-harassment-and-harassment-work-technical-guidance); privacy follows the ICO guidance on monitoring workers (ico.org.uk). The judgement is to write behaviours a colleague could see, not adjectives. "Integrity" stops nobody; "you tell your own manager before you approve anything for someone you are in a relationship with" does.

## Workflow
1. Ask three questions: where your people work (no default country); your values and the situations the code must cover; and who receives disclosures and concerns today.
2. Turn each value into two or three observable behaviours and one example of a breach, marked "Example:" and naming no real person.
3. Write conflicts of interest: what must be disclosed (a close relationship with someone in the reporting line, outside interests), to whom, and what changes after disclosure, for example who now approves pay, reviews and promotion for the other person.
4. Write privacy at work: any recording or monitoring must be necessary, proportionate and disclosed; the specific practice goes to an adviser. Propose no monitoring of individuals.
5. Write the speak-up route: internal and external channels, confidentiality where asked, an anonymous option and its limits, no evidence needed to raise a concern, feedback and timeframes you set. Add the non-retaliation statement and anti-harassment cover, including clients and other third parties.
6. Write what happens after a breach: the concern goes to intake, and conduct matters go to the Disciplinary Procedure. The code states no sanction for any person.

## Output Format
```markdown
# Code of Conduct
Owner: [role] | Version: [n] | Applies in: [country, state] | Review date: [date]
## Our values in practice
| Value | We do | Example of a breach |
|---|---|---|
| [value] | [observable behaviour] | Example: [situation, no names] |
## Relationships and conflicts of interest
| What to disclose | Tell | What changes after |
|---|---|---|
| [relationship or interest] | [role] | [who now approves decisions] |
## Privacy and recording at work
[What is and is not recorded, why, how people are told.] [adviser]
## Speaking up
| Channel | Who | Confidential? | Anonymous? | Feedback by |
|---|---|---|---|---|
| [channel] | [role] | [yes / on request] | [yes, limits] | [timeframe you set] |
## Non-retaliation, harassment and breaches
[Statement that victimisation and harassment, including by third parties, are not accepted.] [adviser]
After a breach: [route to intake, then the disciplinary procedure where conduct is involved].
## Adviser questions
1. [question for a qualified adviser in [country]]
## Decision
[Code owner] approves the code after the adviser answers, by [date]; [leader] signs it and [HR lead] briefs managers by [date].
```

## Done When
- Every value has observable behaviours and a marked example.
- The conflicts section says what changes after disclosure; the speak-up route has channels, confidentiality, anonymity limits and a feedback timeframe.
- Every legal point sits under adviser questions.

## Quality Bar
- No adjective-only values; every line describes something a colleague could see.
- Examples never name or hint at real people or real cases.
- No monitoring proposal targets an individual.
- Claude writes the process, never the verdict: it never screens, ranks, scores or reads people, a named person decides and signs every decision about someone, and every legal point goes to a qualified adviser for the country it concerns.

## Next
Run hr-employee-handbook (Employee Handbook) to add the approved code to the handbook.

## About the makers

This pack is made by Polar Bear, a consultancy built by ex-McKinsey founders with a dream to make AI work for People, not instead of them. We help our clients build people systems and AI-first ways of working, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).