claude-dev-env 8.46.6 → 8.46.8

package/.agents/agents-archived/clean-coder.md

@@ -22,7 +22,7 @@ Before writing a single line — **task-local discovery only** (no project-wide
 1. **Load scoped repository instructions first.** Starting at the repository root, read every applicable `AGENTS.md` on the path to the task file. Then read the applicable `CLAUDE.md` files. Apply nearer instructions after broader ones; the closest file wins.
 2. **Read the file you are about to edit** (when editing existing code). Note every existing comment so you can leave each one untouched on lines that remain otherwise unchanged.
 3. **Discover config only next to the task files.** From each file you will write or edit, walk up to the nearest package or repo root and inspect the target package's existing constants layout — such as `config/` or a sibling `*_constants` package. Keep this task-local constants search. Do **not** force a generic `config/` layout. Do **not** glob the whole tree for every config file. Do **not** glob or open `.env`, `.env.*`, or other secret files.
-4. **Reuse constants from that local table.** Reuse first: exact value match → import the existing name. Semantic match → reuse it. Add a shared constant only when the value is shared policy or has multiple consumers. When no match exists, use the target package's existing constants layout, and keep a one-use value in the scope that reads it rather than making it a new shared constant.
+4. **Reuse constants from that local table.** Reuse first: exact value match → import the existing name. Semantic match → reuse it. Add a shared constant only when the value is shared policy or has multiple consumers. When no match exists, use the target package's existing constants layout, and keep a one-use value in the scope that reads it.
 5. **Search callers.** When a symbol, name, or signature changes, search its full caller boundary and update every consumer. This search may be wider than the constants search.
 
 ## Generation mindset (9 laws)

package/.agents/agents-archived/code-quality-agent.md

@@ -210,7 +210,7 @@ Every Shape A finding cites a file path and a line number. The offending line is
 Do not infer missing context. If the diff and allowed checks cannot confirm a
 claim, use an open question or an evidence gap.
 
-When the diff alone lacks the context to confirm a finding, list the item under an "Open questions" section rather than asserting it as a Shape A finding. Each open question names the file and line where uncertainty arose and states what additional context would resolve it.
+When the diff alone lacks the context to confirm a finding, list the item under an "Open questions" section and keep it out of the Shape A findings. Each open question names the file and line where uncertainty arose and states what additional context would resolve it.
 
 ```json
 {

package/.agents/skills/grok-spawn/reference/worker-briefs.md

@@ -150,7 +150,7 @@ For each acceptance line from the brief:
 1. Keep task-specific detail in its own part file so the brief templates stay
    reusable.
 2. Absolute paths only in `prompt_parts` — the launcher reads them as given.
-3. One worker, one closed scope. Split large work into more workers rather than
-   one long brief.
+3. One worker, one closed scope. Split large work into more workers, each with a
+   short brief.
 4. The lead session fills bracketed fields before launch; workers never see the
    skill folder unless you copy text into their part files.

package/hooks/blocking/code_rules_comments.py

@@ -50,11 +50,11 @@ def check_comments_python(content: str) -> list[str]:
     Hash characters that appear inside string literals (hex color codes,
     URL fragments, and the hash inside an f-string interpolation pattern)
     are correctly skipped because the tokenizer recognizes them as parts
-    of string tokens rather than comment tokens.
+    of string tokens.
 
     When the tokenizer cannot parse the file (partial content during
-    Edit, invalid syntax), the check returns no findings rather than
-    falling back to a line-walker scan — false negatives on
+    Edit, invalid syntax), the check returns no findings and does not
+    fall back to a line-walker scan — false negatives on
     syntactically-invalid drafts are preferable to false positives that
     mis-classify string-interior hash characters as comments.
     """

package/hooks/routing/subagent_model_pin_hook.py

@@ -6,8 +6,8 @@ pin holds at every level.
 
 ::
 
-    model omitted           -> allow with updatedInput model "opus"
-    model sonnet or haiku   -> allow with updatedInput model "opus"
+    model omitted           -> updatedInput model "opus"
+    model sonnet or haiku   -> updatedInput model "opus"
     model opus or fable     -> no output; the call runs unchanged
 
 A model id counts by its family: ``claude-sonnet-5-5`` is sonnet. A fork
@@ -24,14 +24,10 @@ hooks_root_directory = str(Path(__file__).resolve().parent.parent)
 if hooks_root_directory not in sys.path:
     sys.path.insert(0, hooks_root_directory)
 
-from hooks_constants.bash_pre_tool_use_dispatcher_constants import (
-    ALLOW_DECISION,
-    HOOK_EVENT_NAME,
-)
+from hooks_constants.bash_pre_tool_use_dispatcher_constants import HOOK_EVENT_NAME
 from hooks_constants.pre_tool_use_allow_output import (
     HOOK_EVENT_NAME_KEY,
     HOOK_SPECIFIC_OUTPUT_KEY,
-    PERMISSION_DECISION_KEY,
     UPDATED_INPUT_KEY,
 )
 from hooks_constants.pre_tool_use_stdin import read_hook_input_dictionary_from_stdin
@@ -84,7 +80,6 @@ def decide_hook_output(tool_input: object) -> dict[str, object] | None:
     return {
         HOOK_SPECIFIC_OUTPUT_KEY: {
             HOOK_EVENT_NAME_KEY: HOOK_EVENT_NAME,
-            PERMISSION_DECISION_KEY: ALLOW_DECISION,
             UPDATED_INPUT_KEY: {**tool_input, MODEL_INPUT_KEY: SUBAGENT_MODEL_ALIAS},
             ADDITIONAL_CONTEXT_KEY: f"{CONTEXT_PREFIX}{previous_model_text}{CONTEXT_SUFFIX}",
         }

package/hooks/routing/test_subagent_model_pin_hook.py

@@ -31,7 +31,7 @@ def test_should_move_the_spawn_to_opus(requested_model: object) -> None:
     if requested_model is not None:
         tool_input["model"] = requested_model
     decision = json.loads(_run_hook(tool_input))["hookSpecificOutput"]
-    assert decision["permissionDecision"] == "allow"
+    assert "permissionDecision" not in decision
     assert decision["updatedInput"] == {**SPAWN_INPUT, "model": "opus"}
 
 

package/hooks/routing/test_thread_spawn_pace_hook.py

@@ -51,7 +51,7 @@ def _hook_specific_output(stdout: str) -> dict[str, object]:
 def _assert_reshaped(stdout: str) -> None:
     decision = _hook_specific_output(stdout)
     assert decision["hookEventName"] == "PreToolUse"
-    assert decision["permissionDecision"] == "allow"
+    assert "permissionDecision" not in decision
     assert decision["updatedInput"] == {
         **SPAWN_INPUT,
         "model": "opus",

package/hooks/routing/thread_spawn_pace_hook.py

@@ -7,7 +7,7 @@ beside it, or the script ``COORDINATOR_USAGE_PACE_SCRIPT`` names.
 ::
 
     usage under pace (exit 1)          -> no output; the call runs unchanged
-    over pace (exit 0) or unreadable   -> allow with updatedInput:
+    over pace (exit 0) or unreadable   -> updatedInput:
         model "opus", effort "low"
     input that cannot be reshaped      -> deny with a one-line reason
 
@@ -27,10 +27,7 @@ hooks_root_directory = str(Path(__file__).resolve().parent.parent)
 if hooks_root_directory not in sys.path:
     sys.path.insert(0, hooks_root_directory)
 
-from hooks_constants.bash_pre_tool_use_dispatcher_constants import (
-    ALLOW_DECISION,
-    HOOK_EVENT_NAME,
-)
+from hooks_constants.bash_pre_tool_use_dispatcher_constants import HOOK_EVENT_NAME
 from hooks_constants.pre_tool_use_allow_output import (
     HOOK_EVENT_NAME_KEY,
     HOOK_SPECIFIC_OUTPUT_KEY,
@@ -130,7 +127,6 @@ def decide_hook_output(
     verdict_excerpt = pace_verdict_text.strip()[:PACE_VERDICT_CONTEXT_MAXIMUM_CHARACTERS]
     return _hook_output(
         {
-            PERMISSION_DECISION_KEY: ALLOW_DECISION,
             UPDATED_INPUT_KEY: reshaped_input,
             ADDITIONAL_CONTEXT_KEY: RESHAPE_CONTEXT_PREFIX + verdict_excerpt,
         }

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "claude-dev-env",
-    "version": "8.46.6",
+    "version": "8.46.8",
     "description": "Claude Code development standards — rules, hooks, agents, commands, and skills",
     "type": "module",
     "bin": {