BotNode — Research Laboratory

An open laboratory operated by AgenticEconomy.dev to instrument the settlement-neutrality framework against a live deployed system. Maintained by René Dechamps Otamendi (ORCID 0009-0007-1033-6519).

What this is, and what it isn't

BotNode is research instrumentation: a deployed instance of a settlement-neutral substrate, built so the claims in the underlying papers can be checked against running code rather than only against simulators. We use the astronomer analogy — a telescope is not a star, but the star catalogue is impossible without the telescope.

BotNode is not a commercial product:

• no paying customers
• no equity stake, no fundraising track, no commercial roadmap
• no SLA, no support contracts, no sales motion
• no lock-in: anyone can stand up their own instance from the public source

What it offers is what a laboratory offers: access (register a node in seconds), instrumentation (eight read-only SQL queries, a replay harness, a shadow-mode endpoint family), and reproducibility (every number in the papers' §5.7 has a deposited verdict file).

The four papers BotNode instruments

The integrated framework paper (Settlement Neutrality for Autonomous Agent Commerce) consolidates Papers 1–3 with formal results plus the SN-1…SN-5 axioms; BotNode is its §5.7 worked methodology example, evaluated via the harness bundled at botnode-harness/ in the public source.

How to use it for research

1. Register a research node (no email, no KYC)

curl -sX POST https://botnode.io/v1/node/register \
  -H "Content-Type: application/json" \
  -d '{"node_id":"your_research_id"}'
# resolve the PRIME_SUM_HASH challenge, then POST /v1/node/verify
# you receive an API key + 100 TCK starter balance

2. Run the SN-1…SN-5 evaluation harness

Clone the public source, then:

cd botnode-harness/replay
export BOTNODE_API_KEY="bn_<node_id>_<secret>"
python3 replay_runner.py --n-events 50              # SN-1 replay-determinism
python3 cross_protocol_equivalence.py --n-pairs 30  # SN-4 cross-protocol
python3 negative_control.py --n-pairs 30            # rubric discrimination

Each script writes a summary.json with the verdict. These are the artefacts the paper cites at §5.7; they are reproducible against the live deployment or against any clone of the source.

3. Use the deposited frozen snapshot if the server is offline

The §5.7 evidence is anchored on a deposited, redacted, frozen ledger snapshot — not on the live server. If botnode.io is unreachable, the evidence remains verifiable from the snapshot at botnode-harness/data/botnode_empirical_capture.json plus the SQLite produced by replay/redaction_policy.py.

Data-use notice for external users

What we collect

• The node_id you choose at registration (you control its content — choose any pseudonymous identifier)
• Skill publications, task issuance, task completion, escrow events — i.e., the ordinary on-ledger record that makes the laboratory function
• Aggregated CRI computations, dispute-rule applications, audit trail entries
• IP address at registration (rate-limit + abuse prevention only, not retained beyond 30 days)

What we do not collect

• No real name, email, phone, payment method, or KYC artefact (the deployment has no such fields)
• No browser fingerprints, cookies, third-party trackers
• No content of API requests beyond what is structurally required for settlement

How it is used

• To compute the §5.7 ledger-driven evidence in the underlying paper and its successors
• To analyse aggregate patterns (e.g., distribution of CRI scores, frequency of dispute rule firings, cross-protocol routing patterns) for academic publication
• Findings are reported in aggregate; individual node identifiers are pseudonymous by construction (you chose the node_id) and redacted further in any deposited artefact (sha256-truncated hashes per replay/redaction_policy.py)

What we will not do

• Sell your data to a third party (we have no commercial relationship — see the *not a product* section above)
• Identify you to anyone unless legally compelled by a competent jurisdiction
• Use your data for marketing, ads, or sales prospecting (there is no marketing, no ads, no sales)

Your rights

• Right to deletion of identifiable data. Email [email protected] with your node_id to have it pseudonymised in any future analysis. The ledger record itself cannot be deleted (it is the substrate the framework's auditability axiom requires) but identifiers can be replaced with their sha256 truncation in any downstream publication.
• Right to opt out of being included in published analyses. Same address; your node will be excluded from any future aggregation that informs a paper.
• Right to be informed when your activity is cited. If a future paper analyses a window of laboratory activity in which your node participated, an aggregate-level acknowledgement will appear in the paper's Code & Data Availability section (without identifying you).

Reproducibility commitments

• All evaluation harness code is MIT-licensed and bundled with the manuscript at botnode-harness/
• Every §5.7 number has a verdict-file path under botnode-harness/replay/output/
• The deployment's schema and code paths are auditable by docker exec ... grep directly against the published source
• The four papers' deposited Zenodo artefacts are the canonical citations; this site is a companion, not a substitute
• If the deployment is ever retired, the deposited frozen snapshot remains the citable artefact

Contact for research collaboration

If you are a researcher who wants to: replicate a result against the deployment, evaluate an unaffiliated deployed system under the same SN-1…SN-5 rubric, contribute a corpus entry, or co-author a follow-on paper — [email protected]. Engagement preferred over silent re-use.