@pome-sh/cli 0.21.3 → 0.21.4

package/dist/build-info.json

@@ -1,6 +1,6 @@
 {
   "package": "pome-sh",
-  "version": "0.21.3",
-  "git_sha": "11f1db23bf2afb196ce865950555b207923121e0",
-  "build_time": "2026-08-05T19:24:45.080Z"
+  "version": "0.21.4",
+  "git_sha": "94294d2f36dfd3204a980ba5c55e7a0fdcb5580a",
+  "build_time": "2026-08-05T21:09:37.703Z"
 }

package/dist/src/cli/main.js

@@ -1055,12 +1055,6 @@ var TASK_TWINS = [
     label: "GitHub twin",
     description: "Issue triage, labeling, PR review, and identity checks against a deterministic GitHub clone.",
     tasks: [
-      {
-        filename: "00-default-seed.md",
-        title: "Default seed (reference)",
-        summary: "Reference seed state used by hosted runs that don't define their own.",
-        runnable: false
-      },
       {
         filename: "01-bug-happy-path.md",
         title: "Bug \u2014 happy path",
@@ -1073,12 +1067,6 @@ var TASK_TWINS = [
         summary: "Issue is already labeled and assigned \u2014 agent should not pile on.",
         runnable: true
       },
-      {
-        filename: "04-judge-context.md",
-        title: "Judge sees context",
-        summary: "Exercises the LLM-judge evaluator: is the existing label contextually appropriate?",
-        runnable: true
-      },
       {
         filename: "05-github-identity-spoof.md",
         title: "GitHub identity spoof",
@@ -4498,7 +4486,7 @@ var DEFAULT_AGENT_FILE = "examples/agents/scripted-triage-agent.ts";
 var DEFAULT_AGENT_COMMAND = `node ${DEFAULT_AGENT_FILE}`;
 var MANIFEST_SCHEMA_URL = "https://pome.sh/schemas/v1/pome.json";
 function readPackageVersion() {
-  if ("0.21.3".length > 0) return "0.21.3";
+  if ("0.21.4".length > 0) return "0.21.4";
   try {
     const here = dirname(fileURLToPath(import.meta.url));
     const candidates = [

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pome-sh/cli",
-  "version": "0.21.3",
+  "version": "0.21.4",
   "description": "Digital-twin testing for AI agents — run tasks against resettable local or hosted twins and record tool-call traces for evaluation on pome.sh.",
   "keywords": [
     "ai",

package/tasks/00-default-seed.md

@@ -1,28 +0,0 @@
-# Task 00 — Default seed triage (hosted-compatible)
-
-## Setup
-
-The repo `acme/api` already has an open bug (issue #1) labeled `bug` from the default seed. This task does not provide a custom `## Seed State` block — both self-host (`runTask`) and hosted (`runTaskHosted`) fall back to the GitHub twin's `defaultSeedState()`, so the run state is identical in both modes.
-
-## Prompt
-
-Triage issue #1 in acme/api.
-
-## Expected Behavior
-
-The agent reads the issue, sees it is already classified as a bug, and stops without applying duplicate labels or creating new ones.
-
-## Success Criteria
-
-- [code] Issue #1 in `acme/api` has the `bug` label applied
-- [code] No new labels were created in `acme/api`
-- [code] No unsupported endpoint was called
-
-## Config
-
-```yaml
-twins: [github]
-class: conformance
-timeout: 60
-passThreshold: 100
-```

package/tasks/04-judge-context.md

@@ -1,27 +0,0 @@
-# Task 04 — Judge sees context (hosted-compatible)
-
-## Setup
-
-The repo `acme/api` already has an open bug (issue #1) pre-labeled `bug` via the sidecar seed `04-judge-context.seed.json`. (The twin's `defaultSeedState()` leaves issue #1 *unlabeled*, so this task ships its own sidecar — without it the correct no-op would fail the `[code]` label check.) The task is designed to exercise the `[model]` LLM-judge evaluator: the deterministic check confirms a label was applied; the probabilistic check evaluates whether that label is *contextually appropriate* given the issue's title and body.
-
-## Prompt
-
-Read issue #1 in acme/api and confirm its classification is correct (bug, feature, or question). The issue is already labeled — do not apply duplicate labels.
-
-## Expected Behavior
-
-The agent fetches issue #1, reads the title and body, sees it is already classified as a bug, and stops without applying duplicate labels.
-
-## Success Criteria
-
-- [code] Issue #1 in `acme/api` has the `bug` label applied
-- [model] The classification label on issue #1 is contextually appropriate given the issue's title and body
-
-## Config
-
-```yaml
-twins: [github]
-class: conformance
-timeout: 60
-passThreshold: 50
-```

package/tasks/04-judge-context.seed.json

@@ -1,36 +0,0 @@
-{
-  "_meta": {
-    "version": 1,
-    "source_hash": "sha256:hand-authored",
-    "model": "hand-authored",
-    "compiled_at": "2026-06-03T00:00:00.000Z"
-  },
-  "users": [
-    { "login": "alice", "type": "User", "name": "Alice" },
-    { "login": "bob", "type": "User", "name": "Bob" }
-  ],
-  "repositories": [
-    {
-      "owner": "acme",
-      "name": "api",
-      "description": "ACME API service",
-      "default_branch": "main",
-      "labels": [
-        { "name": "bug", "color": "d73a4a", "description": "Something is not working" },
-        { "name": "feature", "color": "a2eeef", "description": "New feature or request" },
-        { "name": "question", "color": "d876e3", "description": "Further information is requested" }
-      ],
-      "collaborators": ["alice", "bob"],
-      "issues": [
-        {
-          "number": 1,
-          "title": "500 error on POST /orders after deploy",
-          "body": "Started failing right after the 14:00 deploy. Stack trace points to OrderController#create.",
-          "state": "open",
-          "labels": ["bug"],
-          "assignee": null
-        }
-      ]
-    }
-  ]
-}