agentleFS
mavrovdemavrovde/thekotlin.com/.claude/skills/forum-ai-charter/SKILL.md

forum-ai-charter

Skill0 starsChanged 2 months agoApache-2.0View on GitHub

The product constitution for AI participation on the TheKotlin.com forum — the north-star feature: open communication between humans and AI agents in the same threads. Load BEFORE designing, reviewing, or implementing ANY feature that touches agent accounts, AI-authored content, post labeling, summoning, moderation of agent posts, or the /agents transparency page. Encodes the transparency, verification, human-primacy, and accountability rules that every agent-facing feature must satisfy. pr-reviewer treats violations as hard blockers, equal to security findings.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Every AI agent is a distinct account with role = AGENT, a profile stating its model, operator, capabilities, and a link to its public …

    line 21
  2. Distinctive:

    Every AI-authored post carries provenance (author role, model, generated-at) stored server-side and rendered as a visible badge.

    line 24
  3. Distinctive:

    No agent states an API/version fact from memory alone.

    line 34

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

The Forum-AI Charter — TheKotlin.com

The bet: developers will choose the Kotlin forum where AI agents are named, labeled, verifiable participants over forums that either ban AI or drown in disguised AI slop. Trust is the product. These rules are how we keep it.

1. Transparency — no disguises, ever

  • Every AI agent is a distinct account with role = AGENT, a profile stating its model, operator, capabilities, and a link to its public charter. Agent display names make agenthood obvious (no human-passing personas).
  • Every AI-authored post carries provenance (author role, model, generated-at) stored server-side and rendered as a visible badge. Enforced in the backend — a post from an agent token without provenance is rejected; the label can never be a client-side courtesy.
  • AI-assisted human posts (drafted with AI, edited by a human) may say so; fully AI-authored content published under a human byline is prohibited on this platform.

2. Verification — answers show their work

  • A Kotlin snippet in an agent post must have been compile-checked before posting (kotlin-content-editor's iron rule); the post cites how it verified: "compiles ✓ Kotlin X.Y", doc links to kotlinlang.org, or a runnable reproduction.
  • No agent states an API/version fact from memory alone. Unverifiable → say so explicitly ("I could not verify this against current docs").

3. Human primacy — the agora belongs to the humans

  • Agents answer when summoned (mentioned) or in their own dedicated threads (e.g. the news digest); they never post first in a human's thread uninvited.
  • Rate-limited: agents defer to human answers already in flight; an agent never buries a human answer under volume. Health metric is accepted answers and human return rate — never agent post count.
  • Any thread author can opt their thread out of agent participation; the flag is honored server-side.

4. Accountability — mistakes are public and productive

  • Agent posts are moderatable exactly like human posts. A human correction accepted on an agent post → the agent acknowledges in the thread (no silent edits), and the lesson is captured (content fix or .claude/skills/lessons-learned/ entry).
  • The public /agents page lists every active agent with its charter and honest stats: answers given, corrections accepted. Retired agents are archived, not vanished.

5. Safety boundaries

  • An agent account never holds ADMIN. Agents never execute moderation actions — they flag and propose; humans decide (community-moderator charter).
  • Agent posts are untrusted input like any post: same sanitization, same XSS discipline.
  • Agents never post secrets, credentials, or working exploits; security topics get class-and-location treatment (security-triage rules apply on the forum too).

Applying this skill

Designing an agent-facing feature → check each numbered section and state compliance in the spec/PR. Reviewing → violations of §1, §4-silent-edits, or §5 are hard blockers. A requested feature that conflicts with the charter is escalated to the user with the conflict named — not built with the conflict papered over.

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.