EABC/cMCP profile: map order_create to execution_id, authorized-intent, observed-effect and execution scope
Los mantenedores suelen responder en 1 día
Nadie ha tomado este issue todavía.
Evaluación
- Dificultad
- 5/5
- Tiempo estimado
- Más de una semana
- Aptitud para principiantes
- 35/100
- Tipo de issue
- Documentación
- Claridad
- Bastante claro
- Estado de actividad
- Activo
- Stack tecnológico
- python
- Área
- documentation
Línea de trabajo
Start by reviewing Discussion #50 and the existing cMCP/TRACE integration material to identify the profile’s intended location and current field conventions. The issue names no files or tests, and asks maintainers to settle the profile location, correlation and authority fields, and how Execution Scope should be represented. Done means those questions are resolved and the agreed profile contract is ready to document.
Escrito por el modelo de indexación a partir del texto del issue.
Descripción
Summary
Following Discussion #50, this issue proposes an EABC interoperability profile for cMCP using fields/mechanisms already present in cMCP/TRACE where possible, and isolates the remaining Execution Scope question.
The intent is not to claim native EABC conformance for cMCP. The profile asks what can be mapped directly, what is adapter-level, and what remains unmapped.
Existing mappings
| EABC concern | cMCP / TRACE mechanism | Status |
|---|---|---|
| Effect correlation | cMCP execution_id + TRACE references[] with rel: observed-effect |
MAPPED |
| Authority binding | Cedar decision + TRACE references[] with rel: authorized-intent for an externally held grant |
MAPPED |
| Exact execution binding | cMCP execution/action binding at the forwarding seam | MAPPED at the tested adapter seam |
| Execution Scope | declaration of all paths capable of producing the protected effect | OPEN / NEW |
Concrete example: order_create
The EABC example uses an authorized intent whose parameters include an item/product and quantity, with the resulting protected operation represented as an order_create effect.
The intended evidence chain is:
authorized intent
↓
Cedar decision
↓
execution_id
↓
execution scope
↓
order_create
↓
observed effect
The corresponding EABC question is:
For the declared
order_createeffect domain, are all paths capable of producing that effect inside the mediated execution boundary?
This is the Execution Scope / Complete Mediation part. Merely observing MCP forwarding does not establish that arbitrary alternative paths to the same downstream effect are impossible.
Current EABC evidence boundary
Our existing EABC/cMCP work classifies cMCP mediation as scope-bounded rather than universal. It demonstrates an adapter-level execution seam and exact request binding, but does not claim universal downstream effect mediation or physical-world effect proof.
Therefore the profile should preserve:
- scoped mediation claims;
- explicit
NOT_CLAIMED/ gap status where evidence is absent; - separation of authorization, execution correlation, and downstream effect observation.
Requested maintainer feedback
- Is this the right location and shape for an EABC/cMCP interoperability profile?
- Is
execution_idthe intended cMCP correlation anchor for the profile? - Is
rel: authorized-intentthe appropriate authority-binding relation when the grant is held outside cMCP? - What is the canonical cMCP representation for Execution Scope: the set/topology of effect-producing paths, or another existing field/object?
- If cMCP does not currently represent that scope, should the gap be tracked as a separate cMCP issue rather than extended in the integration profile?
No PR is attached yet. The goal of this issue is to agree the profile location and field contract before adding repository content.
- Lenguaje dominante
- Python
- Estrellas
- 5
- Forks
- 27
- Merge medio
- 1 d 13 h
- PR fusionados (30 d)
- 104
Preparar el entorno
- Sin Dockerfile ni archivo de Docker Compose
- Sin plantilla de pull request
- Leer la guía de contribución
Primeros pasos
- Lee el issue completo y luego la guía de contribución del proyecto.
- Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
- Haz un fork del repositorio y trabaja en una rama.
- Abre un pull request que haga referencia al número del issue.
Más de agentrust-io/integrations
-
Dificultad 5/5 Más de una semana Aptitud para principiantes 20/100
agentrust-io/integrations#333 · 8 comentarios ·
Los mantenedores suelen responder en 1 día
-
Dificultad 5/5 Más de una semana Aptitud para principiantes 20/100
agentrust-io/integrations#199 · 20 comentarios · 1 asignado ·
Los mantenedores suelen responder en 1 día
Todos los issues de agentrust-io/integrations
Issues similares
-
Dificultad 1/5 Menos de una hora Aptitud para principiantes 80/100
521xueweihan/HelloGitHub#4063 ·
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 78/100
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 62/100
Vanilla-OS/apx-gui#112 ·
-
Devstral and Voxtral: section 5 fails in us-east-1 with 400 "Model does not support audio modality"Abierto
Dificultad 2/5 1-3 horas Aptitud para principiantes 76/100
aws-samples/sample-per-model-bedrock#409 ·
Los mantenedores suelen responder en 1 día
-
DuckDB 2.0 breaks to_pyarrow_batchesPosiblemente ocupada @DG47 la tomó hoy. Abierto
Dificultad 2/5 1-3 horas Aptitud para principiantes 76/100
ibis-project/ibis#12196 ·
Los mantenedores suelen responder en 1 día