Skip to main content
Glama
halla-ai
by halla-ai

ax-commons

Open contribution loop for public AI-adoption programs.

Using an AI tool should be enough to measure progress and to contribute back. ax-commons combines a local MCP server, bilingual task skills, a consented contribution flow and an aggregate-only metrics receiver, so that:

  • the activity logger never captures conversation text,

  • only small-cell-suppressed counts and user-approved contributions are shared,

  • failures count as contributions, and contributors are recognized.

Version 0.2 implements the local MCP logger, explicit public contribution approval and withdrawal, suppressed monthly metrics, and a bilingual web form/dashboard. Real-user installation, pilot measurements, partner approval and outside-reader review remain deferred. See the design, installation guide, and operator handover.

No sharing occurs merely because a server is configured. Preview and approve each contribution; approve each monthly metrics upload. Public copies and Git history cannot be recalled, and a merged contribution withdrawal remains pending until the removal PR is merged and verified.

License

This repository uses two licenses, split by content type:

  • Code under the MIT License: packages/, web/, scripts/, CI configuration, and any other source code.

  • Text and content under CC BY 4.0: commons/ (skills, prompts, cases, issue reports), docs/, and other prose.

If a file's license is unclear from its location, the notice in the file wins. Contributions are accepted under the license of the directory they land in; see CONTRIBUTING.md.

Related MCP Connectors

Related MCP Servers

  • F
    license
    Not graded
    quality
    D
    maintenance
    Captures AI-assisted work into a searchable ledger and exposes it via MCP for querying past conversations and decisions.
    -
  • A
    license
    Not graded
    quality
    F
    maintenance
    Provides agent-native analytics and logging for coding agents, enabling product analytics, error logging, and deploy attribution through an MCP handoff.
    0
    2
    MIT
  • A
    license
    A
    quality
    A
    maintenance
    Enables local, offline AI-text detection over MCP with read-only tools for status checking and statistical analysis of text using calibrated compact models.
    2
    BSD 3-Clause
  • A
    license
    A
    quality
    B
    maintenance
    Provides a local MCP interface to query captured, PII-scrubbed prompt telemetry from coding agents and browser chats, enabling time summaries, prompt search, and activity analysis without data leaving the machine.
    9
    MIT