john/prompts
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

intent adjustments

Browse Source
This commit is contained in:
John Lancaster
2026-06-19 00:01:52 -05:00
parent 59c638c634
commit 07475f972f
2 changed files with 10 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/architecture.md
+8 -4
View File
@@ -21,7 +21,7 @@ This architecture keeps authored content human-friendly while preserving machine
The architecture is designed to satisfy three long-term requirements: The architecture is designed to satisfy three long-term requirements:
1. Methodology must be editable as markdown by humans. 1. Methodology must be editable as markdown by humans.
2. Agents must consume stable, discoverable resource contracts. 2. Agents must consume stable, discoverable resource contracts, with a minimal read-only catalog tool fallback for constrained clients.
3. Public documentation must be pre-built static output served from the application runtime without a separate docs service. 3. Public documentation must be pre-built static output served from the application runtime without a separate docs service.
## System Model ## System Model
@@ -36,7 +36,7 @@ The document resource returns canonical Markdown, while clients can perform any
### Catalog Module ### Catalog Module
The catalog is the canonical discovery layer and publishes normalized records for all modules. The catalog is the canonical discovery layer and publishes normalized records for all modules. It may also expose a minimal set of read-only discovery tools that resolve back to the same canonical markdown content when a client chat surface does not expose MCP resource attachment.
Typical catalog resources: Typical catalog resources:
@@ -128,7 +128,7 @@ Markdown remains easy to review, while contracts remain stable for clients.
### Client Independence ### Client Independence
Clients can use Ask, Edit, or Agent modes without requiring server-owned prompt orchestration. Clients can use Ask, Edit, or Agent modes without requiring server-owned prompt orchestration. However, MCP affordances are still chat-surface-dependent: some clients or sessions expose resource attachment directly, while others make tool invocation the more reliable retrieval path.
## Authoring and Publishing Lifecycle ## Authoring and Publishing Lifecycle
@@ -148,9 +148,13 @@ In-scope:
Out-of-scope: Out-of-scope:
1. Prompt-first orchestration as the primary interface 1. Prompt-first orchestration as the primary interface
2. Large tool inventories duplicating static guidance 2. Large tool inventories duplicating static guidance across skill modules
3. Separate dynamic docs service at runtime 3. Separate dynamic docs service at runtime
Allowed exception:
1. A small catalog-level tool layer is acceptable when it improves client interoperability without creating a second source of truth for skill content.
## Example Content Inputs ## Example Content Inputs
Existing markdown reference sets are valid examples of authored source material for this architecture: Existing markdown reference sets are valid examples of authored source material for this architecture:
docs/mcp_layout.md
+2
View File
@@ -108,6 +108,8 @@ Example mapping model:
Catalog resources provide discovery metadata and stable identifiers. Catalog resources provide discovery metadata and stable identifiers.
When clients cannot attach MCP resources directly, catalog-level tools may retrieve the same underlying skill documents indirectly. This does not create a second content source; it is only an alternate access path to the same markdown-backed contract.
## Why This Pattern ## Why This Pattern
### Operational Simplicity ### Operational Simplicity