fix: allow Claude to chain skills for hook execution (#2227)

* fix: allow Claude to chain skills for hook execution (#2178)

- Set disable-model-invocation to false so Claude can invoke extension
  skills (e.g. speckit-git-feature) from within workflow skills
- Inject dot-to-hyphen normalization note into Claude SKILL.md hook
  sections so the model maps extension.yml command names to skill names
- Replace Unicode checkmark with ASCII [OK] in auto-commit scripts to
  fix PowerShell encoding errors on Windows
- Move Claude-specific frontmatter injection to ClaudeIntegration via
  post_process_skill_content() hook on SkillsIntegration, wired through
  presets and extensions managers
- Add positive and negative tests for all changes

Fixes #2178

* refactor: address PR review feedback

- Preserve line-ending style (CRLF/LF) in _inject_hook_command_note
  instead of always inserting \n, matching the convention used by other
  injection helpers in the same module.

- Extract duplicated _post_process_skill() from extensions.py and
  presets.py into a shared post_process_skill() function in agents.py.
  Both modules now import and call the shared helper.

* fix: match full hook instruction line in regex

The regex in _inject_hook_command_note only matched lines ending
immediately after 'output the following', but the actual template
lines continue with 'based on its `optional` flag:'. Use [^\r\n]*
to capture the rest of the line before the EOL.

* refactor: use integration object directly for post_process_skill_content

Instead of a free function in agents.py that re-resolves the
integration by key, callers in extensions.py and presets.py now
resolve the integration once via get_integration() and call
integration.post_process_skill_content() directly. The base
identity method lives on SkillsIntegration.

src/specify_cli/agents.py

@@ -317,11 +317,6 @@ class CommandRegistrar:
                 "source": source,
             },
         }
-        if agent_name == "claude":
-            # Claude skills should be user-invocable (accessible via /command)
-            # and only run when explicitly invoked (not auto-triggered by the model).
-            skill_frontmatter["user-invocable"] = True
-            skill_frontmatter["disable-model-invocation"] = True
         return skill_frontmatter
 
     @staticmethod

src/specify_cli/extensions.py

@@ -850,6 +850,7 @@ class ExtensionManager:
 
         from . import load_init_options
         from .agents import CommandRegistrar
+        from .integrations import get_integration
         import yaml
 
         written: List[str] = []
@@ -860,6 +861,7 @@ class ExtensionManager:
         if not isinstance(selected_ai, str) or not selected_ai:
             return []
         registrar = CommandRegistrar()
+        integration = get_integration(selected_ai)
 
         for cmd_info in manifest.commands:
             cmd_name = cmd_info["name"]
@@ -939,6 +941,10 @@ class ExtensionManager:
                 f"# {title_name} Skill\n\n"
                 f"{body}\n"
             )
+            if integration is not None and hasattr(integration, "post_process_skill_content"):
+                skill_content = integration.post_process_skill_content(
+                    skill_content
+                )
 
             skill_file.write_text(skill_content, encoding="utf-8")
             written.append(skill_name)

src/specify_cli/integrations/base.py

@@ -1102,6 +1102,16 @@ class SkillsIntegration(IntegrationBase):
             invocation = f"{invocation} {args}"
         return invocation
 
+    def post_process_skill_content(self, content: str) -> str:
+        """Post-process a SKILL.md file's content after generation.
+
+        Called by external skill generators (presets, extensions) to let
+        the integration inject agent-specific frontmatter or body
+        transformations.  The default implementation returns *content*
+        unchanged.  Subclasses may override — see ``ClaudeIntegration``.
+        """
+        return content
+
     def setup(
         self,
         project_root: Path,

src/specify_cli/integrations/claude/__init__.py

@@ -5,11 +5,21 @@ from __future__ import annotations
 from pathlib import Path
 from typing import Any
 
+import re
+
 import yaml
 
 from ..base import SkillsIntegration
 from ..manifest import IntegrationManifest
 
+# Note injected into hook sections so Claude maps dot-notation command
+# names (from extensions.yml) to the hyphenated skill names it uses.
+_HOOK_COMMAND_NOTE = (
+    "- When constructing slash commands from hook command names, "
+    "replace dots (`.`) with hyphens (`-`). "
+    "For example, `speckit.git.commit` → `/speckit-git-commit`.\n"
+)
+
 # Mapping of command template stem → argument-hint text shown inline
 # when a user invokes the slash command in Claude Code.
 ARGUMENT_HINTS: dict[str, str] = {
@@ -148,6 +158,43 @@ class ClaudeIntegration(SkillsIntegration):
             out.append(line)
         return "".join(out)
 
+    @staticmethod
+    def _inject_hook_command_note(content: str) -> str:
+        """Insert a dot-to-hyphen note before each hook output instruction.
+
+        Targets the line ``- For each executable hook, output the following``
+        and inserts the note on the line before it, matching its indentation.
+        Skips if the note is already present.
+        """
+        if "replace dots" in content:
+            return content
+
+        def repl(m: re.Match[str]) -> str:
+            indent = m.group(1)
+            instruction = m.group(2)
+            eol = m.group(3)
+            return (
+                indent
+                + _HOOK_COMMAND_NOTE.rstrip("\n")
+                + eol
+                + indent
+                + instruction
+                + eol
+            )
+
+        return re.sub(
+            r"(?m)^(\s*)(- For each executable hook, output the following[^\r\n]*)(\r\n|\n|$)",
+            repl,
+            content,
+        )
+
+    def post_process_skill_content(self, content: str) -> str:
+        """Inject Claude-specific frontmatter flags and hook notes."""
+        updated = self._inject_frontmatter_flag(content, "user-invocable")
+        updated = self._inject_frontmatter_flag(updated, "disable-model-invocation", "false")
+        updated = self._inject_hook_command_note(updated)
+        return updated
+
     def setup(
         self,
         project_root: Path,
@@ -155,7 +202,7 @@ class ClaudeIntegration(SkillsIntegration):
         parsed_options: dict[str, Any] | None = None,
         **opts: Any,
     ) -> list[Path]:
-        """Install Claude skills, then inject user-invocable, disable-model-invocation, and argument-hint."""
+        """Install Claude skills, then inject Claude-specific flags and argument-hints."""
         created = super().setup(project_root, manifest, parsed_options, **opts)
 
         # Post-process generated skill files
@@ -173,11 +220,7 @@ class ClaudeIntegration(SkillsIntegration):
             content_bytes = path.read_bytes()
             content = content_bytes.decode("utf-8")
 
-            # Inject user-invocable: true (Claude skills are accessible via /command)
-            updated = self._inject_frontmatter_flag(content, "user-invocable")
-
-            # Inject disable-model-invocation: true (Claude skills run only when invoked)
-            updated = self._inject_frontmatter_flag(updated, "disable-model-invocation")
+            updated = self.post_process_skill_content(content)
 
             # Inject argument-hint if available for this skill
             skill_dir_name = path.parent.name  # e.g. "speckit-plan"

src/specify_cli/presets.py

@@ -707,6 +707,7 @@ class PresetManager:
 
         from . import SKILL_DESCRIPTIONS, load_init_options
         from .agents import CommandRegistrar
+        from .integrations import get_integration
 
         init_opts = load_init_options(self.project_root)
         if not isinstance(init_opts, dict):
@@ -716,6 +717,7 @@ class PresetManager:
             return []
         ai_skills_enabled = bool(init_opts.get("ai_skills"))
         registrar = CommandRegistrar()
+        integration = get_integration(selected_ai)
         agent_config = registrar.AGENT_CONFIGS.get(selected_ai, {})
         # Native skill agents (e.g. codex/kimi/agy/trae) materialize brand-new
         # preset skills in _register_commands() because their detected agent
@@ -789,6 +791,10 @@ class PresetManager:
                     f"# Speckit {skill_title} Skill\n\n"
                     f"{body}\n"
                 )
+                if integration is not None and hasattr(integration, "post_process_skill_content"):
+                    skill_content = integration.post_process_skill_content(
+                        skill_content
+                    )
 
                 skill_file = skill_subdir / "SKILL.md"
                 skill_file.write_text(skill_content, encoding="utf-8")
@@ -816,6 +822,7 @@ class PresetManager:
 
         from . import SKILL_DESCRIPTIONS, load_init_options
         from .agents import CommandRegistrar
+        from .integrations import get_integration
 
         # Locate core command templates from the project's installed templates
         core_templates_dir = self.project_root / ".specify" / "templates" / "commands"
@@ -824,6 +831,7 @@ class PresetManager:
             init_opts = {}
         selected_ai = init_opts.get("ai")
         registrar = CommandRegistrar()
+        integration = get_integration(selected_ai) if isinstance(selected_ai, str) else None
         extension_restore_index = self._build_extension_skill_restore_index()
 
         for skill_name in skill_names:
@@ -877,6 +885,10 @@ class PresetManager:
                     f"# Speckit {skill_title} Skill\n\n"
                     f"{body}\n"
                 )
+                if integration is not None and hasattr(integration, "post_process_skill_content"):
+                    skill_content = integration.post_process_skill_content(
+                        skill_content
+                    )
                 skill_file.write_text(skill_content, encoding="utf-8")
                 continue
 
@@ -906,6 +918,10 @@ class PresetManager:
                     f"# {title_name} Skill\n\n"
                     f"{body}\n"
                 )
+                if integration is not None and hasattr(integration, "post_process_skill_content"):
+                    skill_content = integration.post_process_skill_content(
+                        skill_content
+                    )
                 skill_file.write_text(skill_content, encoding="utf-8")
             else:
                 # No core or extension template — remove the skill entirely