Skip to content

Conversation

@brawlaphant
Copy link

This PR adds two small, docs-only coordination artifacts to help humans and agents
review and contribute to agentic-tokenomics more efficiently.

What’s included:

  • docs/CONTRIBUTOR_NAV.md: where different kinds of work should live + PR sizing guidance
  • docs/DEPENDENCIES.md: cross-repo references (koi-research, heartbeat, claude tooling)
  • docs/architecture/STACK_MAP.mmd + .md: a visitable system map rendered by GitHub

Notes:

  • Docs-only change; no mechanism or code behavior changes
  • Intended to reduce review friction for large or multi-agent PRs

…sts)

This commit packages the m010 Reputation Signal as a Phase 1, advisory-only
mechanism that is fully reviewable and runnable offline.

It adds:
- a tightened spec, glossary, invariants, and mechanism-local changelog
- canonical JSON schemas with explicit versioning
- a reference implementation for KPI and score computation
- golden test vectors and deterministic replay fixtures
- a seed-based fixture generator
- a single verify command for WG review (no network calls)

The intent is to reduce restart and review cost for future Regen AI work by
providing concrete, testable artifacts rather than forum-level discussion.
No enforcement, gating, or on-chain actions are implied.
Adds a short dependency map and contributor navigation to clarify repo structure,
cross-repo references, and PR sizing expectations. Docs-only change.
Adds a visitable Mermaid diagram and short explainer showing how
agentic-tokenomics relates to koi-research, heartbeat, and claude tooling.
Docs-only change.
@gemini-code-assist
Copy link
Contributor

Summary of Changes

Hello @brawlaphant, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request significantly expands the project's documentation, providing a foundational understanding of the Regen Network Agentic Tokenomics & Governance System. It outlines the system's architecture, governance models, and detailed specifications for various tokenomic mechanisms and agentic workflows. The added documentation aims to enhance clarity for contributors and reviewers, facilitating more efficient development and collaboration.

Highlights

  • Comprehensive Documentation Added: This pull request introduces a substantial amount of new documentation covering the design, architecture, and implementation phases of the Regen Network Agentic Tokenomics & Governance System. This includes detailed specifications for token utility mechanisms, agentic workflows, governance processes, and system architecture.
  • Contributor Guidance and System Map: New documents provide clear guidance for contributors and a detailed architectural map, aiming to streamline the review process and improve overall efficiency for both human and AI agents involved in the project.
  • Detailed Mechanism Specifications: Specific mechanisms like the 'm010 Reputation Signal' are thoroughly documented, including their purpose, scoring functions, attack models, and integration points, along with reference implementations and test vectors.
  • Phased Development Overview: The documentation is structured into distinct phases (Discovery & Analysis, Mechanism Design & Specification, Implementation & Testing), offering a clear roadmap and detailed insights into each stage of the project's development.
Changelog
  • README_SNIPPET.md
    • Added a contributing section, directing users to docs/CONTRIBUTOR_NAV.md.
  • agentic-tokenomics-main 2/README.md
    • Added a comprehensive overview of the Regen Network Agentic Tokenomics & Governance System.
  • agentic-tokenomics-main 2/WG_PACKET.md
    • Added a packet detailing the m010 Reputation Signal mechanism, its scope, location, and what it enables.
  • agentic-tokenomics-main 2/mechanisms/m010-reputation-signal/README.md
    • Added a summary of the m010 Reputation Signal (v0 advisory), explaining its purpose, outputs, limitations, and how to reference it.
  • agentic-tokenomics-main 2/mechanisms/m010-reputation-signal/SPEC.md
    • Added a detailed specification for the m010 Reputation Signal, covering problem, actors, signal definition, evidence inputs, scoring function, economic linkage, on-chain vs off-chain boundary, attack model, integration points, acceptance tests, and rollout plan.
  • agentic-tokenomics-main 2/phase-1/1.1-stakeholder-value-flow.md
    • Added a document mapping stakeholder ecosystems and value flows within Regen Network.
  • agentic-tokenomics-main 2/phase-1/1.2-tokenomic-mechanisms.md
    • Added an inventory of existing and proposed tokenomic mechanisms within Regen Network.
  • agentic-tokenomics-main 2/phase-1/1.3-agentic-services.md
    • Added a document identifying and evaluating services for autonomous agents within Regen Network.
  • agentic-tokenomics-main 2/phase-1/1.4-governance-architecture.md
    • Added a document defining a 4-layer governance architecture for the system.
  • agentic-tokenomics-main 2/phase-1/1.5-system-architecture.md
    • Added a document synthesizing the system architecture for the Regen Agentic Tokenomics platform.
  • agentic-tokenomics-main 2/phase-1/README.md
    • Added an overview for Phase 1: Discovery & Analysis, summarizing its sub-phases, key findings, and outputs.
  • agentic-tokenomics-main 2/phase-2/2.1-token-utility-mechanisms.md
    • Added protocol-level specifications for 5 token utility mechanisms.
  • agentic-tokenomics-main 2/phase-2/2.2-agentic-workflows.md
    • Added a specification defining 12 agentic workflows across 4 agent personas.
  • agentic-tokenomics-main 2/phase-2/2.3-governance-processes.md
    • Added a specification formalizing governance processes for the agentic tokenomics system.
  • agentic-tokenomics-main 2/phase-2/2.4-agent-orchestration.md
    • Added a specification defining the technical architecture for orchestrating autonomous agents.
  • agentic-tokenomics-main 2/phase-2/2.5-data-schema-integration.md
    • Added a specification defining data schemas, API contracts, and integration patterns.
  • agentic-tokenomics-main 2/phase-2/README.md
    • Added an overview for Phase 2: Mechanism Design & Specification, summarizing its sub-phases and key outputs.
  • agentic-tokenomics-main 2/phase-3/3.1-smart-contract-specs.md
    • Added implementation-ready specifications for smart contracts supporting the agentic tokenomics system.
  • agentic-tokenomics-main 2/phase-3/3.2-agent-implementation.md
    • Added an implementation guide for building autonomous agents using ElizaOS runtime.
  • agentic-tokenomics-main 2/phase-3/3.3-testing-plan.md
    • Added a document specifying testing strategies, test cases, and validation procedures for the agentic tokenomics system.
  • agentic-tokenomics-main 2/phase-3/3.4-security-framework.md
    • Added a document defining the security framework for the agentic tokenomics system, including threat models and audit procedures.
  • agentic-tokenomics-main 2/phase-3/3.5-technical-docs.md
    • Added comprehensive technical documentation for developers, operators, and integrators.
  • agentic-tokenomics-main 2/phase-3/README.md
    • Added an overview for Phase 3: Implementation & Testing, summarizing its sub-phases and key outputs.
  • agentic-tokenomics-main 3/README.md
    • Added a comprehensive overview of the Regen Network Agentic Tokenomics & Governance System (content identical to agentic-tokenomics-main 2/README.md).
  • agentic-tokenomics-main 3/mechanisms/m010-reputation-signal/README.md
    • Added a summary of the m010 Reputation Signal (v0 advisory) (content identical to agentic-tokenomics-main 2/mechanisms/m010-reputation-signal/README.md).
  • agentic-tokenomics-main 3/mechanisms/m010-reputation-signal/SPEC.md
    • Added a detailed specification for the m010 Reputation Signal (content identical to agentic-tokenomics-main 2/mechanisms/m010-reputation-signal/SPEC.md).
  • agentic-tokenomics-main 3/mechanisms/m010-reputation-signal/datasets/README.md
    • Added a README explaining the purpose and usage of m010 replay datasets.
  • agentic-tokenomics-main 3/mechanisms/m010-reputation-signal/datasets/fixtures/v0_sample.json
    • Added a sample JSON fixture for m010 replay datasets.
  • agentic-tokenomics-main 3/mechanisms/m010-reputation-signal/datasets/schema.json
    • Added a JSON schema for m010 replay datasets.
  • agentic-tokenomics-main 3/phase-1/1.1-stakeholder-value-flow.md
    • Added a document mapping stakeholder ecosystems and value flows (content identical to agentic-tokenomics-main 2/phase-1/1.1-stakeholder-value-flow.md).
  • agentic-tokenomics-main 3/phase-1/1.2-tokenomic-mechanisms.md
    • Added an inventory of existing and proposed tokenomic mechanisms (content identical to agentic-tokenomics-main 2/phase-1/1.2-tokenomic-mechanisms.md).
  • agentic-tokenomics-main 3/phase-1/1.3-agentic-services.md
    • Added a document identifying and evaluating services for autonomous agents (content identical to agentic-tokenomics-main 2/phase-1/1.3-agentic-services.md).
  • agentic-tokenomics-main 3/phase-1/1.4-governance-architecture.md
    • Added a document defining a 4-layer governance architecture (content identical to agentic-tokenomics-main 2/phase-1/1.4-governance-architecture.md).
  • agentic-tokenomics-main 3/phase-1/1.5-system-architecture.md
    • Added a document synthesizing the system architecture (content identical to agentic-tokenomics-main 2/phase-1/1.5-system-architecture.md).
  • agentic-tokenomics-main 3/phase-1/README.md
    • Added an overview for Phase 1: Discovery & Analysis (content identical to agentic-tokenomics-main 2/phase-1/README.md).
  • agentic-tokenomics-main 3/phase-2/2.1-token-utility-mechanisms.md
    • Added protocol-level specifications for 5 token utility mechanisms (content identical to agentic-tokenomics-main 2/phase-2/2.1-token-utility-mechanisms.md).
  • agentic-tokenomics-main 3/phase-2/2.2-agentic-workflows.md
    • Added a specification defining 12 agentic workflows across 4 agent personas (content identical to agentic-tokenomics-main 2/phase-2/2.2-agentic-workflows.md).
  • agentic-tokenomics-main 3/phase-2/2.3-governance-processes.md
    • Added a specification formalizing governance processes (content identical to agentic-tokenomics-main 2/phase-2/2.3-governance-processes.md).
  • agentic-tokenomics-main 3/phase-2/2.4-agent-orchestration.md
    • Added a specification defining the technical architecture for orchestrating autonomous agents (content identical to agentic-tokenomics-main 2/phase-2/2.4-agent-orchestration.md).
  • agentic-tokenomics-main 3/phase-2/2.5-data-schema-integration.md
    • Added a specification defining data schemas, API contracts, and integration patterns (content identical to agentic-tokenomics-main 2/phase-2/2.5-data-schema-integration.md).
  • agentic-tokenomics-main 3/phase-2/README.md
    • Added an overview for Phase 2: Mechanism Design & Specification (content identical to agentic-tokenomics-main 2/phase-2/README.md).
  • agentic-tokenomics-main 3/phase-3/3.1-smart-contract-specs.md
    • Added implementation-ready specifications for smart contracts (content identical to agentic-tokenomics-main 2/phase-3/3.1-smart-contract-specs.md).
  • agentic-tokenomics-main 3/phase-3/3.2-agent-implementation.md
    • Added an implementation guide for building autonomous agents (content identical to agentic-tokenomics-main 2/phase-3/3.2-agent-implementation.md).
  • agentic-tokenomics-main 3/phase-3/3.3-testing-plan.md
    • Added a document specifying testing strategies, test cases, and validation procedures (content identical to agentic-tokenomics-main 2/phase-3/3.3-testing-plan.md).
  • agentic-tokenomics-main 3/phase-3/3.4-security-framework.md
    • Added a document defining the security framework (content identical to agentic-tokenomics-main 2/phase-3/3.4-security-framework.md).
  • agentic-tokenomics-main 3/phase-3/3.5-technical-docs.md
    • Added comprehensive technical documentation (content identical to agentic-tokenomics-main 2/phase-3/3.5-technical-docs.md).
  • agentic-tokenomics-main 3/phase-3/README.md
    • Added an overview for Phase 3: Implementation & Testing (content identical to agentic-tokenomics-main 2/phase-3/README.md).
  • agentic-tokenomics-main 4/README.md
    • Added a comprehensive overview of the Regen Network Agentic Tokenomics & Governance System (content identical to agentic-tokenomics-main 2/README.md).
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/README.md
    • Added a summary of the m010 Reputation Signal (v0 advisory), including replay datasets and reference implementation details.
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/SPEC.md
    • Added a detailed specification for the m010 Reputation Signal (content identical to agentic-tokenomics-main 2/mechanisms/m010-reputation-signal/SPEC.md).
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/datasets/README.md
    • Added a README explaining the purpose and usage of m010 replay datasets (content identical to agentic-tokenomics-main 3/mechanisms/m010-reputation-signal/datasets/README.md).
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/datasets/fixtures/v0_sample.json
    • Added a sample JSON fixture for m010 replay datasets (content identical to agentic-tokenomics-main 3/mechanisms/m010-reputation-signal/datasets/fixtures/v0_sample.json).
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/datasets/schema.json
    • Added a JSON schema for m010 replay datasets (content identical to agentic-tokenomics-main 3/mechanisms/m010-reputation-signal/datasets/schema.json).
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/reference-impl/README.md
    • Added a README for the m010 reference implementation, detailing inputs and outputs for canonical computation.
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/reference-impl/m010_kpi.js
    • Added a JavaScript file for computing m010 Key Performance Indicators.
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/reference-impl/m010_score.js
    • Added a JavaScript file for computing the m010 reputation score.
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/reference-impl/test_vectors/vector_v0_sample.expected.json
    • Added an expected output JSON file for m010 test vectors.
  • agentic-tokenomics-main 4/mechanisms/m010-reputation-signal/reference-impl/test_vectors/vector_v0_sample.input.json
    • Added an input JSON file for m010 test vectors.
  • agentic-tokenomics-main 4/phase-1/1.1-stakeholder-value-flow.md
    • Added a document mapping stakeholder ecosystems and value flows (content identical to agentic-tokenomics-main 2/phase-1/1.1-stakeholder-value-flow.md).
  • agentic-tokenomics-main 4/phase-1/1.2-tokenomic-mechanisms.md
    • Added an inventory of existing and proposed tokenomic mechanisms (content identical to agentic-tokenomics-main 2/phase-1/1.2-tokenomic-mechanisms.md).
  • agentic-tokenomics-main 4/phase-1/1.3-agentic-services.md
    • Added a document identifying and evaluating services for autonomous agents (content identical to agentic-tokenomics-main 2/phase-1/1.3-agentic-services.md).
  • agentic-tokenomics-main 4/phase-1/1.4-governance-architecture.md
    • Added a document defining a 4-layer governance architecture (content identical to agentic-tokenomics-main 2/phase-1/1.4-governance-architecture.md).
  • agentic-tokenomics-main 4/phase-1/1.5-system-architecture.md
    • Added a document synthesizing the system architecture (content identical to agentic-tokenomics-main 2/phase-1/1.5-system-architecture.md).
  • agentic-tokenomics-main 4/phase-1/README.md
    • Added an overview for Phase 1: Discovery & Analysis (content identical to agentic-tokenomics-main 2/phase-1/README.md).
  • agentic-tokenomics-main 4/phase-2/2.1-token-utility-mechanisms.md
    • Added protocol-level specifications for 5 token utility mechanisms (content identical to agentic-tokenomics-main 2/phase-2/2.1-token-utility-mechanisms.md).
  • agentic-tokenomics-main 4/phase-2/2.2-agentic-workflows.md
    • Added a specification defining 12 agentic workflows across 4 agent personas (content identical to agentic-tokenomics-main 2/phase-2/2.2-agentic-workflows.md).
  • agentic-tokenomics-main 4/phase-2/2.3-governance-processes.md
    • Added a specification formalizing governance processes (content identical to agentic-tokenomics-main 2/phase-2/2.3-governance-processes.md).
  • agentic-tokenomics-main 4/phase-2/2.4-agent-orchestration.md
    • Added a specification defining the technical architecture for orchestrating autonomous agents (content identical to agentic-tokenomics-main 2/phase-2/2.4-agent-orchestration.md).
  • agentic-tokenomics-main 4/phase-2/2.5-data-schema-integration.md
    • Added a specification defining data schemas, API contracts, and integration patterns (content identical to agentic-tokenomics-main 2/phase-2/2.5-data-schema-integration.md).
  • agentic-tokenomics-main 4/phase-2/README.md
    • Added an overview for Phase 2: Mechanism Design & Specification (content identical to agentic-tokenomics-main 2/phase-2/README.md).
  • agentic-tokenomics-main 4/phase-3/3.1-smart-contract-specs.md
    • Added implementation-ready specifications for smart contracts (content identical to agentic-tokenomics-main 2/phase-3/3.1-smart-contract-specs.md).
  • agentic-tokenomics-main 4/phase-3/3.2-agent-implementation.md
    • Added an implementation guide for building autonomous agents (content identical to agentic-tokenomics-main 2/phase-3/3.2-agent-implementation.md).
  • agentic-tokenomics-main 4/phase-3/3.3-testing-plan.md
    • Added a document specifying testing strategies, test cases, and validation procedures (content identical to agentic-tokenomics-main 2/phase-3/3.3-testing-plan.md).
  • agentic-tokenomics-main 4/phase-3/3.4-security-framework.md
    • Added a document defining the security framework (content identical to agentic-tokenomics-main 2/phase-3/3.4-security-framework.md).
  • agentic-tokenomics-main 4/phase-3/3.5-technical-docs.md
    • Added comprehensive technical documentation (content identical to agentic-tokenomics-main 2/phase-3/3.5-technical-docs.md).
  • agentic-tokenomics-main 4/phase-3/README.md
    • Added an overview for Phase 3: Implementation & Testing (content identical to agentic-tokenomics-main 2/phase-3/README.md).
Activity
  • The pull request was created by brawlaphant.
  • The pull request introduces extensive new documentation for the project's design and implementation phases.
  • The changes are focused on improving contributor understanding and review efficiency through detailed specifications and architectural maps.
Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request adds a significant amount of documentation for the agentic tokenomics system. However, there seems to be an issue with the file structure. The PR description mentions adding a few specific documentation files, but the changes include a large number of files under directories named agentic-tokenomics-main 2, agentic-tokenomics-main 3, and agentic-tokenomics-main 4. These directories appear to contain duplicated or versioned content, which makes the PR difficult to review and could lead to confusion. It would be great if you could clarify the intended file structure and remove any unintended or duplicated files.

I have reviewed the content of the added files (focusing on the agentic-tokenomics-main 2 and agentic-tokenomics-main 4 directories as representative samples) and have left some comments on areas that could be improved for clarity, consistency, and correctness. Please apply these suggestions to all relevant files if the duplication is intentional.

Comment on lines +519 to +520
let lock_expires = env.block.time.plus_days(att_type.lock_period_days);
let challenge_closes = env.block.time.plus_days(att_type.challenge_window_days);
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

critical

The cosmwasm_std::Timestamp type does not have a plus_days method. You should use plus_seconds and calculate the duration in seconds to avoid a compilation error.

Suggested change
let lock_expires = env.block.time.plus_days(att_type.lock_period_days);
let challenge_closes = env.block.time.plus_days(att_type.challenge_window_days);
let lock_expires = env.block.time.plus_seconds(att_type.lock_period_days * 86400);
let challenge_closes = env.block.time.plus_seconds(att_type.challenge_window_days * 86400);

Comment on lines +56 to +58
- if completeness >= 0.9 AND risk < 0.3: recommend_approve (confidence: 0.95)
- if completeness >= 0.7 AND risk < 0.5: recommend_conditional (confidence: 0.75)
- else: recommend_reject (confidence: 0.85)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The decision logic here is ambiguous. It's not clear if these are sequential if statements or an if-else if-else chain. For example, a case with completeness = 0.95 and risk = 0.4 would be rejected by the first rule, but could be conditionally approved by the second. Please clarify the evaluation order. Using an if/else if/else structure would make the logic unambiguous.

For example:

    - if completeness >= 0.9 AND risk < 0.3: recommend_approve (confidence: 0.95)
    - else if completeness >= 0.7 AND risk < 0.5: recommend_conditional (confidence: 0.75)
    - else: recommend_reject (confidence: 0.85)

This would ensure that cases like completeness = 0.95, risk = 0.4 are handled as recommend_conditional instead of falling through to recommend_reject.

Comment on lines +75 to +79
## Mechanisms

- [m010 Reputation Signal (v0 advisory)](mechanisms/m010-reputation-signal/SPEC.md)


Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

This 'Mechanisms' section appears to be redundant, as the same information is provided under the 'Key Specifications' -> 'Mechanisms (canonical specs)' section below (line 87). Removing this duplicate section would improve the document's clarity.

Comment on lines +118 to +150
```
regen-agentic-tokenomics/
├── README.md
├── mechanisms
│ └── m010-reputation-signal
│ ├── README.md
│ └── SPEC.md
├── phase-1/
│ ├── README.md
├── mechanisms
│ └── m010-reputation-signal
│ ├── README.md
│ └── SPEC.md
│ ├── 1.1-stakeholder-value-flow.md
│ ├── 1.2-tokenomic-mechanisms.md
│ ├── 1.3-agentic-services.md
│ ├── 1.4-governance-architecture.md
│ └── 1.5-system-architecture.md
├── phase-2/
│ ├── README.md
├── mechanisms
│ └── m010-reputation-signal
│ ├── README.md
│ └── SPEC.md
│ ├── 2.1-token-utility-mechanisms.md
│ ├── 2.2-agentic-workflows.md
│ ├── 2.3-governance-processes.md
│ ├── 2.4-agent-orchestration.md
│ └── 2.5-data-schema-integration.md
├── phase-3/
├── phase-4/
└── phase-5/
```
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The repository structure diagram is malformed. It contains duplicated mechanisms entries and incorrect indentation for phase-specific files, which makes the structure difficult to understand. A corrected structure would improve clarity. For example:

regen-agentic-tokenomics/
├── README.md
├── mechanisms/
│   └── m010-reputation-signal/
│       ├── README.md
│       └── SPEC.md
├── phase-1/
│   ├── README.md
│   ├── 1.1-stakeholder-value-flow.md
│   ├── 1.2-tokenomic-mechanisms.md
│   ├── 1.3-agentic-services.md
│   ├── 1.4-governance-architecture.md
│   └── 1.5-system-architecture.md
├── phase-2/
│   ├── README.md
│   ├── 2.1-token-utility-mechanisms.md
│   ├── 2.2-agentic-workflows.md
│   ├── 2.3-governance-processes.md
│   ├── 2.4-agent-orchestration.md
│   └── 2.5-data-schema-integration.md
├── phase-3/
├── phase-4/
└── phase-5/

{
"id": "M003",
"name": "Transaction Fees",
"module": "x/auth",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

There's an inconsistency for mechanism M003. The table on line 31 lists the module as x/auth, x/bank, but the JSON here only lists x/auth. For consistency, this should be updated to include x/bank as well.

Suggested change
"module": "x/auth",
"module": "x/auth, x/bank",

@@ -0,0 +1,29 @@
import { median } from "./m010_kpi.js";
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The median function is imported but not used in this file. It should be removed to keep the code clean.

@brawlaphant
Copy link
Author

Clarification on scope:

Although this PR branch contains multiple legacy upload artifact folders, the only intended changes in this PR are:

  • docs/CONTRIBUTOR_NAV.md
  • docs/DEPENDENCIES.md
  • docs/architecture/STACK_MAP.md
  • docs/architecture/STACK_MAP.mmd

The duplicate agentic-tokenomics-main* folders are upload artifacts from earlier work and are NOT intended as canonical or new documentation. Please focus review only on the docs/ additions.

Thank you.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant