developer-community-moderation skill (samber): code of conduct and moderation

Writes a developer community's code of conduct and the private moderation runbook that makes it enforceable.

Installs: 12.4K; on skills.sh
Stars: 7; on GitHub
Files: 5; besides SKILL.md
Updated: Nov 2026; last repo commit

developer-community-moderation writes the two deliverables that make behavioural policy real: a published code of conduct and the private runbook behind it. It first decides which of four jobs you have (a live incident, no code yet, an unenforceable code, or moderation without structure), then maps scope, designs a four-rung enforcement ladder, builds reporting channels, staffs the team, writes the runbook and sets platform controls. A publish gate of eleven items must pass before anything ships.

It triggers on "a code of conduct", "CODE_OF_CONDUCT.md", "banning a member", "harassment" or "our Discord is getting out of hand". It bundles no scripts. SKILL.md says to stabilize a live incident first, and to name the gap if the policy would not be enforced against the founder or top maintainer.

Good for

Maintainers and community managers with no code of conduct yet
Teams handling a conduct report right now
Communities flooded by spam or AI-generated submissions

Install developer-community-moderation

Claude Code

Copy this prompt.

Install the developer-community-moderation skill in Claude Code. Run this command:

/plugin marketplace add samber/developer-relations-skills
/plugin install developer-relations-skills@developer-relations-skills

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

The README recommends installing the whole collection because its skills reference each other, and developer-community-moderation names other skills in the collection to hand work to. The whole-collection routes add all 93 skills in the repo (counted from the git tree); the repo holds no hooks, agents, commands or MCP server files, and 14 of its skills bundle a script. Pick one route per agent.

From the project README. Source.

What's in the skill

When it loads

Writes a developer community's code of conduct and the moderation playbook behind it - scope, enforcement ladder, reporting channels, incident-response runbook, moderator roster, platform controls. Use whenever the user mentions a code of conduct, CODE_OF_CONDUCT.md, community moderation, moderator recruitment, an escalation ladder, banning or suspending a member, harassment, trolls, brigading, spam or AI-slop floods, or a conduct report they need to handle - including vaguer phrasings like "our Discord is getting out of hand", "we need community rules", or "someone reported a maintainer". Covers company-run and volunteer open-source communities. Do NOT use to pick the platform - use samber/developer-relations-skills@developer-community-launch.

Sections in SKILL.md

Detect the request first
Interview
Workflow
The ladder is a gate, the outcomes are a menu
Deliverables
Invocation examples
Scope and jurisdiction
Enforcement ladder
Reporting
Response team
Runbook summary
Platform controls
Publish gate
Open questions for the user
Community types that change the answer
Responses that make things worse

Key facts

Author: Samuel Berthe (named in SKILL.md metadata, LICENSE and the plugin manifest)
License: MIT (LICENSE file, SKILL.md frontmatter, plugin.json and marketplace.json agree)
Skill version: 1.0.4 (SKILL.md metadata)
Installs on skills.sh: 12,410 (Nov 2, 2026)
GitHub stars: 8 for the whole samber/developer-relations-skills repo (Nov 2, 2026)
Last repo commit: Nov 1, 2026
Audits on skills.sh: Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
What it includes: 6 files in the git tree: SKILL.md, evals/evals.json and 4 files under references/. No scripts.
Needs: SKILL.md lists nothing to install
Part of: The developer-relations-skills plugin: 93 skill folders in the git tree
Works in: The README documents installs for Claude Code, Claude.ai, Codex, Cursor and Gemini CLI, all for the whole set

What it includes

Sorts the request into four jobs, each with a different first move.
Detect the request
Asks 11 questions; questions 2 and 3 decide what may be drafted.
Asks 11 questions; question 2 (the body above the moderators) and 3 (enforcement against the powerful) decide what may be drafted.
Interview
Runs twelve steps from scope and jurisdiction to a review loop.
Twelve steps from scope and jurisdiction to a review loop and persisted decisions.
Workflow
Uses four rungs with trigger and consequence, plus non-ladder outcomes.
Four rungs with trigger and consequence, plus non-ladder outcomes such as content removal and systemic fixes.
Enforcement ladder
Sets up a private channel, a backup path and a no-retaliation rule.
A private channel, a second path for reports about the first recipient, a stated acknowledgment target and a no-retaliation rule.
Reporting channels
Ranks rate limits, entry gates, trust levels, filters and bans by inflow stopped.
Rate limits, entry gates, trust levels, filters and bans, ranked by inflow stopped per moderator hour.
Platform controls
Writes the code, a private runbook, a public page, a roster and a control sheet.
Code of conduct, private runbook, public what-to-expect page, moderator roster, control sheet and announcement.
Deliverables
Requires eleven checks, such as the code applying to the most powerful people.
Eleven checks, such as the code applying to the community's most powerful people.
Publish gate
Lists five responses that make things worse, such as requiring an apology.
Five responses that make things worse, such as requiring an apology. SKILL.md says Aurora and Gardiner name four of them, and tells Claude to keep those four out of the ladder and the runbook.
Responses that make things worse

From the official docs

Permissions

Medium risk

developer-community-moderation is policy-writing instructions with no scripts, and it may save notes to devrel-context.md.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

Conduct reports name real people
The incident runbook asks for the name, handle, role and affiliation of the people involved, and a live case puts allegations about them into the conversation. SKILL.md's memory instruction stores task and project context in devrel-context.md when memory lives in a file, and does not say to keep names out of it. The skill bundles no scripts and names no network access, so the medium rating rests on the sensitivity of that material. Our advice: share the minimum a case needs and keep case files out of public repositories. github.com
It may save notes to devrel-context.md
SKILL.md tells Claude to keep task and project context in devrel-context.md when memory lives in a file, or in the memory system your tool already uses. Check the file before you commit it. github.com
The whole-set install adds 93 skills, 14 with scripts
The README says to install every skill. The git tree has 93 skill folders, and 14 of them hold .py or .sh files. This skill bundles no scripts, and the tree holds no hooks, agents or MCP config files. github.com

Technical details

Turning on read-only	SKILL.md describes no read-only mode.
Sign-in and keys	SKILL.md names no account, key or token.
Scopes	Claude writes a code of conduct and runbook in the conversation Conduct reports, names and evidence you share go into the conversation Writes devrel-context.md when memory lives in a file

Questions

What does developer-community-moderation do?

It writes a code of conduct and the moderation playbook that enforces it. You get an enforcement ladder, reporting channels with a backup path, a private runbook, a moderator roster and platform control settings.

How do I install it?

Use the install box at the top. The README says to install every skill in the repo, because the skills point to each other. The box shows the whole-set route for each app.

What if there is an incident happening right now?

SKILL.md treats it as job 1: skip document work, stabilize first by checking everyone is safe, preserving evidence, using the platform control that stops the harm and acknowledging the reporter. Policy is written afterwards from what the case showed.

Will it write a code of conduct my founder is exempt from?

SKILL.md says not to. It cites Aurora and Gardiner's rule that a community must not adopt a code of conduct that will not apply to its most powerful people, and offers fixing the exemption or publishing community guidelines without conduct-enforcement language.

Is it safe to install?

The skill bundles no scripts. Reports, names and evidence about community members go into your conversation, so share only what a case needs. The whole-set install adds 93 skills, 14 of which bundle scripts.

How is it different from oss-issue-triage?

This skill handles behaviour and conduct. SKILL.md routes the technical issue workflow of labels, templates and stale policy to samber/developer-relations-skills@oss-issue-triage.

Information

updated Nov 2, 2026

Repository: samber/developer-relations-skills
Folder: skills/developer-community-moderation

samber/developer-relations-skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 4

README · github.com
developer-community-moderation SKILL.md · github.com
LICENSE · github.com
incident-response-runbook.md · github.com