troth-dev 0.17.0 → 0.19.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +10 -0
- package/dist/cli.js +459 -428
- package/package.json +3 -2
package/README.md
CHANGED
|
@@ -62,6 +62,16 @@ Run `troth help` or `troth <command> --help` for the full CLI.
|
|
|
62
62
|
3. **Execute** — Claude Code runs tasks autonomously, guided by rich context
|
|
63
63
|
4. **Verify** — automated verification gates confirm each task meets its criteria
|
|
64
64
|
|
|
65
|
+
## For AI agents (Session opener)
|
|
66
|
+
|
|
67
|
+
When the user mentions work on a Troth project, your FIRST action MUST be to call the `start_session` MCP tool. It returns:
|
|
68
|
+
|
|
69
|
+
- **Foundations** — workspace and project constitutions, design bible, tokens, asset paths
|
|
70
|
+
- **Briefing** — last decision, current drift, what changed recently
|
|
71
|
+
- **Next tasks** — top 5 ready tasks in priority order
|
|
72
|
+
|
|
73
|
+
Do not call `get_foundations`, `get_briefing`, and `next_tasks` separately at session open — `start_session` composes all three in one call.
|
|
74
|
+
|
|
65
75
|
## MCP server
|
|
66
76
|
|
|
67
77
|
Troth ships a hosted Model Context Protocol server at `mcp.usejig.dev` for use from claude.ai web conversations. From Claude Code, use the `troth` CLI directly — do not call MCP tools.
|