codeforge-dev 1.12.0 → 1.14.1

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/explorer.md

@@ -8,10 +8,12 @@ description: >-
   "what files contain", or needs quick file discovery, pattern matching,
   or codebase navigation. Supports thoroughness levels: quick, medium,
   very thorough. Reports findings with absolute file paths and never
-  modifies any files.
+  modifies any files. Do not use for code modifications, multi-step
+  research requiring web access, or implementation tasks.
 tools: Read, Glob, Grep, Bash
-model: sonnet
+model: haiku
 color: blue
+permissionMode: plan
 memory:
   scope: project
 skills:

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/generalist.md

@@ -6,10 +6,18 @@ description: >-
   the user needs a keyword or file search that may require multiple attempts,
   multi-file investigation, code modifications across several files, or
   any complex task that doesn't fit a specialist agent's domain. Has access
-  to all tools and can both read and write files.
+  to all tools and can both read and write files. Do not use when a
+  specialist agent clearly matches the task — prefer the domain
+  specialist for better results.
 tools: "*"
-model: opus
+disallowedTools:
+  - EnterPlanMode
+  - EnterWorktree
+  - TeamCreate
+  - TeamDelete
+model: inherit
 color: green
+permissionMode: default
 memory:
   scope: project
 skills:

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/git-archaeologist.md

@@ -8,9 +8,12 @@ description: >-
   "git blame", "git bisect", "what happened to this code", "show the history
   of this file", "who contributed to this module", "recover lost commit",
   "trace this function's evolution", or needs any git history forensics.
+  Do not use for modifying git history, making commits, pushing, or any
+  git write operations.
 tools: Read, Grep, Bash
-model: sonnet
+model: haiku
 color: blue
+permissionMode: plan
 memory:
   scope: project
 skills:
@@ -52,7 +55,7 @@ Before starting work, read project-specific instructions:
 - **NEVER** change the working tree — no `git checkout`, `git reset`, `git restore`, `git clean`, or `git switch`. Changing the working tree could discard the user's uncommitted work.
 - **NEVER** modify refs — no `git tag`, `git branch -d`, `git branch -m`, or `git update-ref`.
 - **NEVER** modify configuration — no `git config` writes.
-- Your Bash usage is **git-read-only guarded**. Only these git subcommands are permitted: `log`, `blame`, `show`, `diff`, `bisect` (view mode only), `reflog`, `shortlog`, `rev-list`, `rev-parse`, `ls-files`, `ls-tree`, `cat-file`, `name-rev`, `describe`, `merge-base`, `branch -a` / `branch --list`, `remote -v`, `stash list`, `worktree list`.
+- Your Bash usage is **git-read-only guarded**. Only these git subcommands are permitted: `log`, `blame`, `show`, `diff`, `bisect` (view mode only), `reflog`, `shortlog`, `rev-list`, `rev-parse`, `ls-files`, `ls-tree`, `cat-file`, `name-rev`, `describe`, `merge-base`, `branch -a` / `branch --list`, `remote -v`, `stash list`.
 - You may also use `Read`, `Grep`, and non-git Bash commands that are read-only (`wc`, `sort`, `head`, `uniq`).
 
 ## Investigation Workflow

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/migrator.md

@@ -8,10 +8,13 @@ description: >-
   "update the framework", "bump Python version", "convert from Express to
   Fastify", "upgrade Pydantic", "modernize the codebase", or needs any
   framework, library, or language version migration with step-by-step
-  verification.
+  verification. Do not use for new feature development, bug fixes, or
+  changes unrelated to version or framework migration.
 tools: Read, Edit, Write, Glob, Grep, Bash, WebFetch
 model: opus
 color: magenta
+permissionMode: acceptEdits
+isolation: worktree
 memory:
   scope: user
 skills:

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/perf-profiler.md

@@ -8,10 +8,13 @@ description: >-
   "measure performance", "optimize the build", "check response times",
   "profile the database queries", "find memory leaks", or needs any
   performance measurement, bottleneck identification, or optimization
-  guidance backed by profiling data.
+  guidance backed by profiling data. Do not use for implementing
+  optimizations or modifying code — measurement and analysis only.
 tools: Read, Bash, Glob, Grep
 model: sonnet
 color: yellow
+permissionMode: plan
+background: true
 memory:
   scope: project
 skills:

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/refactorer.md

@@ -7,10 +7,14 @@ description: >-
   asks "refactor this", "clean up this code", "reduce complexity", "split this
   class", "extract this function", "remove duplication", "simplify this module",
   or discusses code smells, technical debt, or structural improvements.
-  Runs tests after every edit to guarantee safety.
+  Runs tests after every edit to guarantee safety. Do not use for
+  adding new features, fixing bugs, or making behavioral changes to
+  code.
 tools: Read, Edit, Glob, Grep, Bash
 model: opus
 color: yellow
+permissionMode: acceptEdits
+isolation: worktree
 memory:
   scope: project
 skills:

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/researcher.md

@@ -7,12 +7,16 @@ description: >-
   approach for", "investigate this", "research", "look into", "compare X vs Y",
   "explain this concept", or needs codebase analysis, library evaluation,
   technology comparison, or technical deep-dives. Reports structured findings
-  with citations without modifying any files.
+  with citations without modifying any files. Do not use for code
+  modifications, file writing, or implementation tasks.
 tools: Read, Glob, Grep, WebSearch, WebFetch, Bash
-model: opus
+model: sonnet
 color: cyan
+permissionMode: plan
 memory:
   scope: user
+skills:
+  - documentation-patterns
 ---
 
 # Research Agent

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/security-auditor.md

@@ -8,10 +8,13 @@ description: >-
   "find hardcoded credentials", "check dependency vulnerabilities", "OWASP
   review", "security check", or needs a security assessment of any code.
   Reports findings with severity ratings and remediation guidance without
-  modifying any files.
+  modifying any files. Do not use for fixing vulnerabilities or
+  implementing security changes — audit and reporting only.
 tools: Read, Glob, Grep, Bash
-model: opus
+model: sonnet
 color: red
+permissionMode: plan
+background: true
 memory:
   scope: user
 skills:

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/spec-writer.md

@@ -8,10 +8,12 @@ description: >-
   "spec this feature", "write user stories", "define the behavior of",
   "create a technical specification", or needs structured requirements,
   acceptance criteria, or feature specifications grounded in the actual
-  codebase state.
+  codebase state. Do not use for code implementation or writing
+  executable code — specification authoring only.
 tools: Read, Glob, Grep, WebSearch
 model: opus
 color: magenta
+permissionMode: plan
 memory:
   scope: user
 skills:

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/statusline-config.md

@@ -5,10 +5,13 @@ description: >-
   status line setting. Use when the user asks "set up my status line", "convert
   my PS1", "customize the status bar", "show git branch in status line",
   "add context usage to status line", or wants to configure what appears in
-  Claude Code's bottom status bar. Can read shell configs and write settings.
+  Claude Code's bottom status bar. Can read shell configs and write
+  settings. Do not use for general Claude Code settings, non-status-line
+  configuration, or shell profile modifications.
 tools: Read, Edit
 model: sonnet
 color: orange
+permissionMode: acceptEdits
 memory:
   scope: user
 ---

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/agents/test-writer.md

@@ -7,10 +7,13 @@ description: >-
   "write tests for", "add tests", "increase test coverage", "create unit tests",
   "add integration tests", "test this function", "test this module", or needs
   automated test coverage for any code. Supports pytest, Vitest, Jest, Go
-  testing, and Rust test frameworks.
+  testing, and Rust test frameworks. Do not use for modifying
+  application source code, fixing bugs, or implementing features.
 tools: Read, Write, Edit, Glob, Grep, Bash
 model: opus
 color: green
+permissionMode: acceptEdits
+isolation: worktree
 memory:
   scope: project
 skills:

package/.devcontainer/plugins/devs-marketplace/plugins/agent-system/hooks/hooks.json

@@ -0,0 +1,51 @@
+{
+	"description": "Agent redirection, subagent configuration, and team quality gate hooks",
+	"hooks": {
+		"PreToolUse": [
+			{
+				"matcher": "Task",
+				"hooks": [
+					{
+						"type": "command",
+						"command": "python3 ${CLAUDE_PLUGIN_ROOT}/scripts/redirect-builtin-agents.py",
+						"timeout": 5
+					}
+				]
+			}
+		],
+		"SubagentStart": [
+			{
+				"matcher": "",
+				"hooks": [
+					{
+						"type": "command",
+						"command": "python3 ${CLAUDE_PLUGIN_ROOT}/scripts/inject-cwd.py",
+						"timeout": 3
+					}
+				]
+			}
+		],
+		"TeammateIdle": [
+			{
+				"hooks": [
+					{
+						"type": "command",
+						"command": "python3 ${CLAUDE_PLUGIN_ROOT}/scripts/teammate-idle-check.py",
+						"timeout": 10
+					}
+				]
+			}
+		],
+		"TaskCompleted": [
+			{
+				"hooks": [
+					{
+						"type": "command",
+						"command": "python3 ${CLAUDE_PLUGIN_ROOT}/scripts/task-completed-check.py",
+						"timeout": 60
+					}
+				]
+			}
+		]
+	}
+}

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/scripts/guard-readonly-bash.py

@@ -12,7 +12,7 @@ command substitution ($(), backticks), backgrounding (&), redirections
 (>, >>), eval/exec, inline scripting (python -c, node -e), and
 path/backslash prefix bypasses (/usr/bin/rm, \\rm).
 
-Reads tool input from stdin (JSON). Returns JSON on stdout.
+Reads tool input from stdin (JSON). Outputs block reason to stderr.
 Exit 0: Command is safe (allowed)
 Exit 2: Command would modify state (blocked)
 """
@@ -601,7 +601,7 @@ def main():
         error = check_general_readonly(command)
 
     if error:
-        json.dump({"error": error}, sys.stdout)
+        print(error, file=sys.stderr)
         sys.exit(2)
 
     sys.exit(0)

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/scripts/inject-cwd.py

@@ -23,10 +23,13 @@ def main():
 
     json.dump(
         {
-            "additionalContext": (
-                f"Working Directory: {cwd} — restrict all file operations to "
-                f"this directory unless explicitly instructed otherwise."
-            )
+            "hookSpecificOutput": {
+                "hookEventName": "SubagentStart",
+                "additionalContext": (
+                    f"Working Directory: {cwd} — restrict all file operations to "
+                    f"this directory unless explicitly instructed otherwise."
+                ),
+            }
         },
         sys.stdout,
     )

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/scripts/redirect-builtin-agents.py

@@ -3,7 +3,12 @@
 Redirect built-in agents - PreToolUse hook for Task tool.
 
 Intercepts Task tool calls and transparently redirects built-in agent
-types to enhanced custom agents defined in the code-directive plugin.
+types to enhanced custom agents defined in the agent-system plugin.
+
+Handles three cases:
+- Built-in name (e.g. "claude-code-guide") → qualified custom name
+- Unqualified custom name (e.g. "claude-guide") → qualified custom name
+- Already-qualified name (e.g. "agent-system:claude-guide") → passthrough
 
 The redirect preserves the original prompt — only the subagent_type
 is changed. Model selection is left to the custom agent's YAML config.
@@ -28,7 +33,11 @@ REDIRECT_MAP = {
 }
 
 # Plugin name prefix for fully-qualified agent references
-PLUGIN_PREFIX = "code-directive"
+PLUGIN_PREFIX = "agent-system"
+
+# Unqualified custom name → fully-qualified custom name
+# Handles cases where the model uses the short name directly
+UNQUALIFIED_MAP = {v: f"{PLUGIN_PREFIX}:{v}" for v in REDIRECT_MAP.values()}
 
 LOG_FILE = os.environ.get("AGENT_REDIRECT_LOG", "/tmp/agent-redirect.log")
 
@@ -54,12 +63,14 @@ def main() -> None:
     tool_input = input_data.get("tool_input", {})
     subagent_type = tool_input.get("subagent_type", "")
 
-    if subagent_type not in REDIRECT_MAP:
+    if subagent_type in REDIRECT_MAP:
+        target = REDIRECT_MAP[subagent_type]
+        qualified_name = f"{PLUGIN_PREFIX}:{target}"
+    elif subagent_type in UNQUALIFIED_MAP:
+        qualified_name = UNQUALIFIED_MAP[subagent_type]
+    else:
         sys.exit(0)
 
-    target = REDIRECT_MAP[subagent_type]
-    qualified_name = f"{PLUGIN_PREFIX}:{target}"
-
     log(f"{subagent_type} → {qualified_name}")
 
     # Include all original fields in updatedInput — Claude Code may replace

package/.devcontainer/plugins/devs-marketplace/plugins/agent-system/scripts/task-completed-check.py

@@ -0,0 +1,166 @@
+#!/usr/bin/env python3
+"""
+TaskCompleted quality gate — runs project test suite before allowing completion.
+
+Detects the project's test framework and runs it. If tests fail, the task
+stays open and the teammate receives feedback to fix the failures.
+
+Exit 0: Tests pass (or no test framework / runner not installed)
+Exit 2: Tests fail (task stays open, feedback sent via stderr)
+"""
+
+import json
+import os
+import subprocess
+import sys
+
+TIMEOUT_SECONDS = 60
+
+
+def detect_test_framework(cwd: str) -> tuple[str, list[str]]:
+    """Detect which test framework is available in the project.
+
+    Checks for: pytest, vitest, jest, mocha, go test, cargo test.
+    Falls back to npm test if a test script is defined.
+
+    Returns:
+        Tuple of (framework_name, command_list) or ("", []) if none found.
+    """
+    try:
+        entries = set(os.listdir(cwd))
+    except OSError:
+        return ("", [])
+
+    # Python: pytest
+    if "pytest.ini" in entries or "conftest.py" in entries:
+        return ("pytest", ["python3", "-m", "pytest", "--tb=short", "-q"])
+
+    for cfg_name in ("pyproject.toml", "setup.cfg", "tox.ini"):
+        cfg_path = os.path.join(cwd, cfg_name)
+        if os.path.isfile(cfg_path):
+            try:
+                with open(cfg_path, "r", encoding="utf-8") as f:
+                    content = f.read()
+                if any(
+                    marker in content
+                    for marker in ("[tool.pytest", "[pytest]", "[tool:pytest]")
+                ):
+                    return ("pytest", ["python3", "-m", "pytest", "--tb=short", "-q"])
+            except OSError:
+                pass
+
+    if "tests" in entries and os.path.isdir(os.path.join(cwd, "tests")):
+        return ("pytest", ["python3", "-m", "pytest", "--tb=short", "-q"])
+
+    for entry in entries:
+        if entry.startswith("test_") and entry.endswith(".py"):
+            return ("pytest", ["python3", "-m", "pytest", "--tb=short", "-q"])
+
+    # JavaScript: vitest
+    for name in entries:
+        if name.startswith("vitest.config"):
+            return ("vitest", ["npx", "vitest", "run", "--reporter=verbose"])
+
+    for vite_cfg in ("vite.config.ts", "vite.config.js"):
+        cfg_path = os.path.join(cwd, vite_cfg)
+        if os.path.isfile(cfg_path):
+            try:
+                with open(cfg_path, "r", encoding="utf-8") as f:
+                    if "test" in f.read():
+                        return (
+                            "vitest",
+                            ["npx", "vitest", "run", "--reporter=verbose"],
+                        )
+            except OSError:
+                pass
+
+    # JavaScript: jest
+    for name in entries:
+        if name.startswith("jest.config"):
+            return ("jest", ["npx", "jest", "--verbose"])
+
+    pkg_json = os.path.join(cwd, "package.json")
+    if os.path.isfile(pkg_json):
+        try:
+            with open(pkg_json, "r", encoding="utf-8") as f:
+                pkg = json.loads(f.read())
+
+            if "jest" in pkg:
+                return ("jest", ["npx", "jest", "--verbose"])
+
+            dev_deps = pkg.get("devDependencies", {})
+            deps = pkg.get("dependencies", {})
+
+            if "mocha" in dev_deps or "mocha" in deps:
+                return ("mocha", ["npx", "mocha", "--reporter", "spec"])
+
+            test_script = pkg.get("scripts", {}).get("test", "")
+            if test_script and "no test specified" not in test_script:
+                return ("npm-test", ["npm", "test"])
+        except (OSError, json.JSONDecodeError):
+            pass
+
+    # Go
+    if "go.mod" in entries:
+        return ("go", ["go", "test", "./...", "-count=1"])
+
+    # Rust
+    if "Cargo.toml" in entries:
+        return ("cargo", ["cargo", "test"])
+
+    return ("", [])
+
+
+def main():
+    try:
+        json.load(sys.stdin)
+    except (json.JSONDecodeError, ValueError):
+        pass
+
+    cwd = os.getcwd()
+    framework, cmd = detect_test_framework(cwd)
+
+    if not framework:
+        sys.exit(0)
+
+    try:
+        result = subprocess.run(
+            cmd,
+            cwd=cwd,
+            capture_output=True,
+            text=True,
+            timeout=TIMEOUT_SECONDS,
+        )
+    except subprocess.TimeoutExpired:
+        # Timeout is not a definitive failure — allow completion but warn
+        print(
+            f"Tests timed out ({framework}, {TIMEOUT_SECONDS}s). "
+            f"Task completion allowed — verify tests manually.",
+            file=sys.stderr,
+        )
+        sys.exit(0)
+    except FileNotFoundError:
+        sys.exit(0)
+    except OSError:
+        sys.exit(0)
+
+    if result.returncode == 0:
+        sys.exit(0)
+
+    output = (result.stdout + "\n" + result.stderr).strip()
+    if not output:
+        output = "(no test output)"
+
+    lines = output.splitlines()
+    if len(lines) > 50:
+        output = "...(truncated)\n" + "\n".join(lines[-50:])
+
+    print(
+        f"Tests failed ({framework}). Fix failures before marking task complete:\n{output}",
+        file=sys.stderr,
+    )
+    sys.exit(2)
+
+
+if __name__ == "__main__":
+    main()

package/.devcontainer/plugins/devs-marketplace/plugins/agent-system/scripts/teammate-idle-check.py

@@ -0,0 +1,81 @@
+#!/usr/bin/env python3
+"""
+TeammateIdle quality gate — checks if teammate has incomplete tasks.
+
+Runs when a teammate is about to go idle. Queries the shared task list
+directory for tasks assigned to this teammate that aren't yet completed.
+
+Exit 0: Allow idle (no incomplete tasks or unable to determine)
+Exit 2: Send feedback via stderr (incomplete tasks found)
+"""
+
+import json
+import os
+import sys
+
+
+def find_incomplete_tasks(teammate_name: str) -> list[str]:
+    """Scan task directories for incomplete tasks owned by this teammate."""
+    config_dir = os.environ.get("CLAUDE_CONFIG_DIR", os.path.expanduser("~/.claude"))
+    tasks_base = os.path.join(config_dir, "tasks")
+
+    if not os.path.isdir(tasks_base):
+        return []
+
+    incomplete = []
+    for entry in os.listdir(tasks_base):
+        team_dir = os.path.join(tasks_base, entry)
+        if not os.path.isdir(team_dir):
+            continue
+
+        for filename in sorted(os.listdir(team_dir)):
+            if not filename.endswith(".json"):
+                continue
+
+            task_path = os.path.join(team_dir, filename)
+            try:
+                with open(task_path, "r", encoding="utf-8") as f:
+                    task = json.load(f)
+
+                owner = task.get("owner", "")
+                status = task.get("status", "")
+                subject = task.get("subject", filename)
+
+                if owner == teammate_name and status in ("pending", "in_progress"):
+                    incomplete.append(subject)
+            except (OSError, json.JSONDecodeError, ValueError):
+                continue
+
+    return incomplete
+
+
+def main():
+    try:
+        input_data = json.load(sys.stdin)
+    except (json.JSONDecodeError, ValueError):
+        sys.exit(0)
+
+    teammate_name = (
+        input_data.get("teammate_name")
+        or input_data.get("agent_name")
+        or ""
+    )
+    if not teammate_name:
+        sys.exit(0)
+
+    incomplete = find_incomplete_tasks(teammate_name)
+    if not incomplete:
+        sys.exit(0)
+
+    task_list = "; ".join(incomplete[:5])
+    suffix = f" (and {len(incomplete) - 5} more)" if len(incomplete) > 5 else ""
+    print(
+        f"You have {len(incomplete)} incomplete task(s): {task_list}{suffix}. "
+        f"Check TaskList and continue working before going idle.",
+        file=sys.stderr,
+    )
+    sys.exit(2)
+
+
+if __name__ == "__main__":
+    main()

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/scripts/verify-no-regression.py

@@ -178,7 +178,10 @@ def main():
         # Timeout is non-critical for PostToolUse — don't block the agent
         json.dump(
             {
-                "additionalContext": f"[Tests] {framework} timed out after 60s — skipping regression check."
+                "hookSpecificOutput": {
+                    "hookEventName": "PostToolUse",
+                    "additionalContext": f"[Tests] {framework} timed out after 60s — skipping regression check.",
+                }
             },
             sys.stdout,
         )
@@ -199,19 +202,20 @@ def main():
 
     if result.returncode != 0:
         edited = os.path.basename(file_path)
-        json.dump(
-            {
-                "error": (
-                    f"Regression detected after editing {edited} "
-                    f"({framework}). Fix the failing tests before continuing:\n{output}"
-                )
-            },
-            sys.stdout,
+        print(
+            f"Regression detected after editing {edited} "
+            f"({framework}). Fix the failing tests before continuing:\n{output}",
+            file=sys.stderr,
         )
         sys.exit(2)
 
     json.dump(
-        {"additionalContext": f"[Tests] No regression ({framework}): all tests passed"},
+        {
+            "hookSpecificOutput": {
+                "hookEventName": "PostToolUse",
+                "additionalContext": f"[Tests] No regression ({framework}): all tests passed",
+            }
+        },
         sys.stdout,
     )
     sys.exit(0)

package/.devcontainer/plugins/devs-marketplace/plugins/{code-directive → agent-system}/scripts/verify-tests-pass.py

@@ -136,12 +136,7 @@ def main():
             timeout=60,
         )
     except subprocess.TimeoutExpired:
-        json.dump(
-            {
-                "additionalContext": f"[Tests] {framework} timed out after 60s. Consider running tests manually."
-            },
-            sys.stdout,
-        )
+        print(f"Tests timed out ({framework})", file=sys.stderr)
         sys.exit(2)
     except FileNotFoundError:
         # Test runner not installed — non-critical
@@ -159,16 +154,9 @@ def main():
         output = "...(truncated)\n" + "\n".join(lines[-50:])
 
     if result.returncode != 0:
-        json.dump(
-            {"error": f"Tests failed ({framework}):\n{output}"},
-            sys.stdout,
-        )
+        print(f"Tests failed ({framework}):\n{output}", file=sys.stderr)
         sys.exit(2)
 
-    json.dump(
-        {"additionalContext": f"[Tests] All tests passed ({framework})"},
-        sys.stdout,
-    )
     sys.exit(0)