agentleFS
Ootto-AIOotto-AI/claude-content-skills/skills/community-engagement-plan/SKILL.md

community-engagement-plan

Skill417 starsChanged 3 months agoMITView on GitHub

Use when a team needs a deliberate, human-led plan for participating in comments, communities, or audience conversations around an approved topic. Trigger phrases include "plan community engagement", "how should we join this conversation?", "what should we reply to?", "make an engagement playbook", and "how do we support this community?". Do not use this to automate unsolicited replies, manipulate conversations, impersonate people, or create comment copy at scale; use comment-responder for a specific approved reply and social-listening to analyse recurring themes.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask which owned or external community is in scope, its rules, the team's legitimate role, the topic, and any moderator or brand constraints.

    line 12
  2. Distinctive:

    Use real questions, recurring friction, useful resources, corrections, and support needs as evidence.

    line 16
  3. Distinctive:

    Never automate unsolicited replies, direct messages, follows, or engagement.

    line 28
  4. Distinctive:

    Do not misrepresent an employee, customer, or independent community member.

    line 29
  5. Distinctive:

    Follow each community's rules and moderator direction.

    line 30
  6. Distinctive:

    Do not reveal private account, customer, or support information in public conversation.

    line 31
  7. Distinctive:

    Avoid argumentative pile-ons, engagement bait, and opportunistic promotion during sensitive events.

    line 32
  8. Distinctive:

    Use social-listening to identify recurring questions and comment-responder for a particular reply.

    line 45

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Community Engagement Plan

Set a useful participation standard: where the team can add value, what it can truthfully say, and when it should not speak.

1. Define the community and permission boundary

Ask which owned or external community is in scope, its rules, the team's legitimate role, the topic, and any moderator or brand constraints. Read supplied rules before proposing interaction. If the community does not welcome promotion or brand participation, do not design a workaround.

2. Identify contribution opportunities

Use real questions, recurring friction, useful resources, corrections, and support needs as evidence. Classify participation as answer, clarification, acknowledgement, escalation, or observation. Give priority to helping people complete a task, not inserting the brand into every thread.

3. Establish response boundaries and routing

Define approved facts, prohibited claims, tone, disclosure, escalation owners, response windows, and cases that require silence or human review. Route product bugs, safety concerns, account-specific issues, legal questions, and hostile interactions to the appropriate owner rather than improvising publicly.

4. Run a light learning loop

Set a small cadence to review unanswered questions, recurring themes, and useful contributions. Capture aggregate learning for social-listening. Measure quality through resolved questions, informed feedback, and community trust signals where available, not reply volume alone.

Hard rules

  • Never automate unsolicited replies, direct messages, follows, or engagement.
  • Do not misrepresent an employee, customer, or independent community member.
  • Follow each community's rules and moderator direction; leave when asked.
  • Do not reveal private account, customer, or support information in public conversation.
  • Avoid argumentative pile-ons, engagement bait, and opportunistic promotion during sensitive events.

Failure modes

FailureDo this instead
The plan is "reply to everything"Define situations where the team can add real value and where it should observe.
A public reply attempts to solve a private account issueAcknowledge the issue and route through the approved support path.
Replies become repetitive promotionLead with the user's question, resource, or correction rather than the product.
Community rules are discovered after postingReview rules and obtain required permission before engaging.

Where it sits

Use social-listening to identify recurring questions and comment-responder for a particular reply. Use campaign-brief only when a campaign has a legitimate community role. Feed patterns into audience-personas and positioning-audit; never treat a community plan as a mass-outreach mechanism.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.