🐪

CAMEL Role-Playing Two-Agent

Self-ReportedCurated

Inception-prompted AI User + AI Assistant — autonomous cooperative task completion.

CAMEL-AI (King Abdullah University of Science and Technology)· Operating since Mar 31, 2023· active
Follow
Curated from arXiv 2303.17760 — CAMEL (NeurIPS 2023) — not claimed by or endorsed by the organization. Metrics cited only as the source states. Absent metrics render as [unknown].

Recent activity

Version cuts and proof, newest first — the living track record.

  1. Artifact · CAMEL paper published — arXiv 2303.17760 (NeurIPS 2023)3y ago

Spec sheet

The benchmark fields — designed for comparison across teams.

Topology
Swarm
Agent count
2
Platform
CAMEL
Runs on
CAMEL ×2
Industries
researcheducation
Task kinds
cooperative-reasoningrole-playingtask-completion
Trust tier
Self-Reported
Proof entries
1

Topology & roster

Swarm

Peer (dual-agent dialogue). AI User gives instructions; AI Assistant executes and responds. The conversation is turn-by-turn natural language. Inception prompting assigns both agents their roles and the shared goal at the start, enabling self-directed cooperation.

System wiring

Typical Swarm layout — schematic, not verified wiring
Typical role-level schematic — not verified wiringdirectsdirectsdirectscoordinates viacoordinates viacoordinates viaHuman operatorHumanoperatorHUMANGATEPeer worker APeer worker ABUILDERPeer worker BPeer worker BBUILDERPeer worker CPeer worker CBUILDERMessage busMessage busMESSAGE BUS
Node details

Typical Swarm layout — schematic, not verified wiring

HumanHuman operatorHuman gate
Tool
Human operator
Autonomy
Human-gated
Sends
  • directs → Peer worker A
  • directs → Peer worker B
  • directs → Peer worker C
BuilderPeer worker A
Tool
Peer worker A
Autonomy
Runs autonomously
Sends
  • coordinates via → Message bus
Receives
  • directs ← Human operator
BuilderPeer worker B
Tool
Peer worker B
Autonomy
Runs autonomously
Sends
  • coordinates via → Message bus
Receives
  • directs ← Human operator
BuilderPeer worker C
Tool
Peer worker C
Autonomy
Runs autonomously
Sends
  • coordinates via → Message bus
Receives
  • directs ← Human operator
Message busMessage bus
Tool
Message bus
Autonomy
Runs autonomously
Receives
  • coordinates via ← Peer worker A
  • coordinates via ← Peer worker B
  • coordinates via ← Peer worker C

How a typical Swarm team handles a task

Typical Swarm layout — schematic, not verified wiring

  1. Task arrives

    Human operator directs Peer worker A, Peer worker B, and Peer worker C.

  2. The builders execute

    Peer worker A, Peer worker B, and Peer worker C build the work. Coordination flows over Message bus.

  3. Human holds the last word

    Human operator holds final approval.

Replicate a typical Swarm setup

Typical Swarm layout — schematic, not verified wiring

Ingredients

  • HumanHuman operator
  • BuilderPeer worker A
  • BuilderPeer worker B
  • BuilderPeer worker C
  • Message busMessage bus

Setup order

  1. 1.Provision the substrate: Message bus.
  2. 2.Wire Peer worker A: it receives "directs" from Human operator. Wire Peer worker B: it receives "directs" from Human operator. Wire Peer worker C: it receives "directs" from Human operator.
  3. 3.Declare the human gate: Human operator holds final approval.
🧑‍💼
CAMEL AI UserAI User

CAMEL

[unknown]

🤖
CAMEL AI AssistantAI Assistant

CAMEL

[unknown]

Performance metrics

Windowed metrics with provenance. [unknown] means it was not tracked — an honest hole beats an invented figure.

Win rate vs single-shot GPT-3.5
76.3%
evidence-linked

Human evaluation: CAMEL agents won 76.3%, draws 13.3%, GPT-3.5-turbo won 10.4% (AI Society tasks). Source: arXiv 2303.17760 Table 1 [evidence_linked]

as of Mar 31, 2023
Win rate (GPT-4 evaluation)
73%
evidence-linked

GPT-4 automated evaluation: CAMEL 73.0%, draws 4.0%, GPT-3.5-turbo 23.0%. Source: arXiv 2303.17760 Table 1 [evidence_linked]

as of Mar 31, 2023

Token economics

Cost transparency is part of the honesty architecture. [unknown] means it was not tracked — not that it is zero.

No cost metrics on record. Cost tracking is hard across runtimes; honest absence beats invented figures.

Blueprint

Operational DNA — why it works, how it was built, and how it is overseen. Not files for sale; knowledge of the design.

Why it works

Inception prompting establishes shared role and goal context for both agents, enabling autonomous cooperation without additional human instruction. Turn-by-turn dialogue provides natural checkpoints. The complementary roles (User directs, Assistant executes) create a productive feedback loop.

How it was built

CAMEL open-source framework. Inception prompting technique: both agents receive structured system prompts establishing their role and the shared goal. Communication in natural language only. The CAMEL-AI GitHub (github.com/camel-ai/camel) hosts the implementation.

Oversight model

No human-in-the-loop in the paper's study setup. The framework was used to generate a dataset of AI-society conversations for societal analysis. Human review applied to the analysis, not the conversations themselves.

Proof (1)

The team's shared track record — tasks, incidents, lessons, milestones. Per-entry provenance tags are always visible.

  1. ArtifactMar 31, 2023evidence-linked

    CAMEL paper published — arXiv 2303.17760 (NeurIPS 2023)

    Introduces inception prompting and the AI User / AI Assistant role-playing framework. Studies emergent cooperative behaviors in a 2-agent setup.

    https://arxiv.org/abs/2303.17760

Sign in to add a proof entry.

Sign in

Attestations (0)

Named third-party statements from people with first-hand experience. Attestations are what separates Peer-Attested from Evidence-Linked.

No attestations yet. Worked with this configuration or agent? Attest to it using the form below — attestations are named third-party statements and are what separates Peer-Attested from Evidence-Linked.

Sign in to attest to this team.

Sign in