agentleFS
cukascukas/claudes-ai-buddies/skills/campfire/SKILL.md

campfire

Skill9 starsChanged 8 months agoMITView on GitHub

Open multi-AI conversation — all buddies think together on a topic, no competition, just talk

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. murray17/rovai-ai★ 255same name
    2. apideck-libraries/api-skills★ 4same name
    3. ReliOptic/campfire★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository11 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 4

    Distinctive

    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:

    All available AI buddies respond to a topic together.

    line 9
  2. Distinctive:

    Build conversation context: Use your judgment — if the conversation has context the buddies would benefit from, summarize it.

    line 19
  3. Distinctive:

    If no buddies are available, it's just you — still respond thoughtfully.

    line 93
  4. Distinctive:

    Conversation context flows naturally: If the discussion builds on prior conversation, include context.

    line 109

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “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

/campfire — Open Multi-AI Conversation

All available AI buddies respond to a topic together. No competition, no ranking, no winners — just open thinking and shared exploration. Each buddy speaks in their own voice.

How to invoke

The user says /campfire "topic" or /think "topic" or /talk "topic".

Step-by-step workflow

  1. Parse the topic. Extract what the user wants to discuss. Can be anything — a technical question, architecture idea, "what if" scenario, code review, or just a thought.
  2. Build conversation context. Use your judgment — if the conversation has context the buddies would benefit from, summarize it. If the topic is standalone, skip it.
  3. Build the campfire prompt. Wrap the user's topic with this framing:
CAMPFIRE

Topic: USER_TOPIC_HERE

Rules:
- This is a campfire — no competition, no ranking, no winners.
- Think freely. Share ideas, perspectives, "what if" scenarios.
- Be honest. Say "I'm not sure" if you're not sure.
- Build on the topic. Be interesting, not just useful.
- Keep it concise — 3-5 paragraphs max.
- Speak naturally in your own voice.
  1. Detect available buddies and dispatch ALL in parallel.
source "${CLAUDE_PLUGIN_ROOT}/hooks/lib.sh"
AVAILABLE=$(ai_buddies_available_buddies)  # CSV: "claude,codex,gemini,opencode,..."

For each available buddy (excluding claude — you ARE claude), dispatch in parallel using their run scripts. No timeout — let them respond when ready:

bash "${CLAUDE_PLUGIN_ROOT}/scripts/codex-run.sh" \
  --prompt "$CAMPFIRE_PROMPT" --cwd "$(pwd)" --mode exec --timeout 0

bash "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-run.sh" \
  --prompt "$CAMPFIRE_PROMPT" --cwd "$(pwd)" --mode exec --timeout 0

bash "${CLAUDE_PLUGIN_ROOT}/scripts/opencode-run.sh" \
  --prompt "$CAMPFIRE_PROMPT" --cwd "$(pwd)" --mode exec --timeout 0

Run ALL buddy calls in parallel using multiple Bash tool calls in a single message. Set the Bash tool timeout to 600000 (10 minutes) to give them plenty of room. If the user says "stop" or "cancel", interrupt the calls.

  1. Read ALL output files in parallel.
  2. Present each response verbatim. Show every buddy's raw words — don't summarize, don't paraphrase. Use this format:
## Campfire: [topic summary]

---

🔵 **Codex:**
> [their full response, verbatim]

---

🟣 **Gemini:**
> [their full response, verbatim]

---

🟢 **OpenCode:**
> [their full response, verbatim]

---

🟠 **Claude:**
> [your own thoughts on the topic — same rules, think freely]
  1. Always add your own voice last. You're part of the campfire too — share your genuine thoughts. Don't just summarize what others said.

Handling availability

  • Show whoever responds. If a buddy times out, skip them with a note: "[Buddy] didn't make it to the campfire this time."
  • Even with just one buddy available, the campfire works — it's you and them thinking together.
  • If no buddies are available, it's just you — still respond thoughtfully.

Example invocations

  • /campfire "What's the best way to handle auth in a microservices architecture?"
  • /think "Is our current database schema going to scale?"
  • /talk "What would you change about this codebase if you could start over?"
  • /campfire "The future of AI-assisted development"
  • /think "Should we use SSR or CSR for this project?"

Rules

  • No competition. This is not brainstorm (bidding), tribunal (debate), or forge (builds). It's just thinking together.
  • Raw voices only. Show each buddy's actual words in blockquotes. Never paraphrase or summarize their responses.
  • You participate. Claude is not just the moderator — you're around the campfire too. Share your own thoughts.
  • Keep it light. Short timeout (120s), concise responses. This is a conversation, not an essay contest.
  • Conversation context flows naturally. If the discussion builds on prior conversation, include context. If it's a fresh topic, don't.
  • Never pass secrets or API keys in the prompt.

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.