@galaxy9day/executor-adapter 0.10.0 → 0.10.2

package/README.md

@@ -221,9 +221,16 @@ When `.trellis/` isn't present, the main orchestrator agent (Claude Code / Codex
 - Runtime files go to `/tmp/executor-adapter/pi-workers/<worker-id>/` instead of `.trellis/.runtime/pi-workers/<worker-id>/`
 - A fingerprint lock keyed on `(scope, extra_instructions)` prevents accidental concurrent identical dispatches
 
-## Trellis sub-agent templates
+## Trellis 0.6 custom agents and skills
 
-This package ships ready-to-copy Claude Code / Trellis custom agent templates in `templates/claude/agents/`:
+Trellis 0.6.x has two extension points that matter here:
+
+- **Custom sub-agents** (`.trellis/agents/*.md` for channel runtime, or platform-specific agent files such as `.claude/agents/*.md`) isolate a role/prompt and can be spawned by the orchestrator.
+- **Custom skills** (`*/skills/<name>/SKILL.md`) are auto-triggered workflow modules. They are the right place for usage policy, but not a replacement for the MCP server itself.
+
+`executor-adapter` remains an MCP server. It is not a native Trellis channel provider, and it does not install or mutate `.trellis/agents` files. Trellis native channel workers should continue to use `.trellis/agents/implement.md` / `check.md` or your own channel agents. Use the templates below when you want a Claude Code sub-agent to call the MCP and summarize the executor result, keeping long dispatch output out of the main context.
+
+This package ships ready-to-copy Claude Code custom agent templates in `templates/claude/agents/`:
 
 ```bash
 cp templates/claude/agents/*.md <your-project>/.claude/agents/
@@ -239,13 +246,14 @@ Task(subagent_type="trellis-pi-check")
 
 This keeps long dispatch output out of the main context. The sub-agent receives the MCP result, reads reports only when needed, and returns a short summary to the orchestrator.
 
-Use `trellis-codex-implement` for the normal GPT implementation path. Use `trellis-pi-implement` only when you explicitly want Pi provider routing for implementation, and `trellis-pi-check` for read-only cross-model review. The Codex template omits `model` so `codex exec` uses this machine's Codex default from `$CODEX_HOME/config.toml`; the Pi templates resolve logical routes from `~/.pi/config.toml`.
+Use `trellis-codex-implement` for the normal GPT implementation path. Use `trellis-pi-implement` only when you explicitly want Pi provider routing for implementation, and `trellis-pi-check` for read-only cross-model review. The templates set `effort: xhigh` for the Claude Code sub-agent itself; the MCP `thinking` parameter separately defaults to `xhigh` for the Codex/Pi executor it launches. The Codex template omits `model` so `codex exec` uses this machine's Codex default from `$CODEX_HOME/config.toml`; the Pi templates resolve logical routes from `~/.pi/config.toml`.
 
-No Trellis `inject-subagent-context` hook edit is required. `dispatch` assembles Trellis task context itself from the task artifacts (`prd.md`, `design.md`, `implement.md`, `implement.jsonl`, and `check.jsonl` where applicable).
+No Trellis `inject-subagent-context` hook edit is required. `dispatch` assembles Trellis task context itself from the task artifacts (`prd.md`, `design.md`, `implement.md`, `implement.jsonl`, and `check.jsonl` where applicable). For Codex/Pi/other platforms, translate these templates to the platform's native custom-agent syntax; do not put MCP tool names in `.trellis/agents/*.md`, because those files are provider-level channel runtime prompts, not MCP-capability declarations.
 
-See the Trellis custom agents docs: https://docs.trytrellis.app/advanced/custom-agents
+See the Trellis docs:
 
-These files follow the Claude Code agent format. Users on other platforms (Cursor, OpenCode, Codex, or similar) should translate the frontmatter and tool declarations to their platform's native agent syntax.
+- Custom agents: https://docs.trytrellis.app/advanced/custom-agents
+- Custom skills: https://docs.trytrellis.app/advanced/custom-skills
 
 ## Forward compatibility
 
@@ -298,7 +306,7 @@ The full dispatch/verification protocol is documented in the executor-adapter sk
 | Node | 20 |
 | `pi` CLI (pi executor) | any recent version, on PATH |
 | `codex` CLI (codex executor) | any version with `codex exec --json` (2025+), on PATH, logged in via `codex login` |
-| Trellis (for channel mode) | `0.6.0-beta.10+` |
+| Trellis (for channel mode) | `0.6.0+` (`0.6.2` tested) |
 | Trellis (for spec assembly) | any 0.5+ |
 
 ## License

package/index.js

@@ -45,7 +45,7 @@ import { EXECUTORS, resolveExecutor } from './executors.js';
 
 const SERVER_NAME = 'executor-adapter';
 const LEGACY_SERVER_NAMES = ['pi-adapter'];
-const SERVER_VERSION = '0.10.0';  // keep in sync with package.json
+const SERVER_VERSION = '0.10.2';  // keep in sync with package.json
 const TMP_RUNTIME_DIR = path.join(os.tmpdir(), SERVER_NAME);
 const CHANNEL_ENV_ALIASES = ['TRELLIS_CHANNEL', 'TRELLIS_CHANNEL_NAME'];
 const TRELLIS_BIN_ENV = 'TRELLIS_BINARY';

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@galaxy9day/executor-adapter",
-  "version": "0.10.0",
+  "version": "0.10.2",
   "description": "Multi-executor MCP (Codex CLI for GPT implementation, Pi CLI for cross-model review) for Trellis-aware and standalone use. Reads Trellis task artifacts, runs the executor in an isolated subprocess/worktree, and can optionally emit non-invasive Trellis channel audit messages.",
   "type": "module",
   "main": "index.js",
@@ -15,7 +15,7 @@
   "dependencies": {
     "@modelcontextprotocol/sdk": "^1.29.0",
     "@iarna/toml": "^2.2.5",
-    "@mindfoldhq/trellis-core": "^0.6.0-beta.13"
+    "@mindfoldhq/trellis-core": "^0.6.2"
   },
   "engines": {
     "node": ">=20"

package/templates/claude/agents/trellis-codex-implement.md

@@ -5,6 +5,7 @@ description: >-
   executor via executor-adapter in an isolated worktree, returning only a short
   summary with result_class, changed files, and apply command. Use this as the
   normal GPT implementation path for Trellis tasks.
+effort: xhigh
 tools: mcp__executor-adapter__dispatch, mcp__executor-adapter__preview_prompt, mcp__executor-adapter__read_report, mcp__executor-adapter__smoke, Read, Grep, Glob, Bash
 ---
 
@@ -73,4 +74,3 @@ main orchestrator context. Dispatch only when the task is ready and bounded.
 - Do not trust executor exit code 0 without checking the auto-validation
   result.
 - If auto-validation fails, report the failure as a blocker or follow-up need.
-*** End Patch

package/templates/claude/agents/trellis-pi-check.md

@@ -4,6 +4,7 @@ description: >-
   Cross-model code review via the Pi reviewer model through executor-adapter; read-only;
   supplements, but does not replace, the orchestrator's own verification and
   native trellis-check. Returns severity-graded findings only.
+effort: xhigh
 tools: mcp__executor-adapter__dispatch, mcp__executor-adapter__read_report, Read, Grep, Glob, Bash
 ---
 

package/templates/claude/agents/trellis-pi-implement.md

@@ -4,6 +4,7 @@ description: >-
   Dispatch mechanical, well-specified implementation tasks to the Pi executor via
   the executor-adapter MCP only when provider routing outside Codex is explicitly
   needed. For normal GPT implementation, use trellis-codex-implement instead.
+effort: xhigh
 tools: mcp__executor-adapter__dispatch, mcp__executor-adapter__preview_prompt, mcp__executor-adapter__read_report, mcp__executor-adapter__smoke, Read, Grep, Glob, Bash
 ---