fix-hook-errors
Systematic debugging workflow for resolving Claude Code plugin hook loading failures. Use when /doctor reports "Hook load failed" with a Zod validation error for any plugin. Covers the three most common failure modes: wrong schema shape (array vs record), stale plugin cache not updated, and duplicate hooks declaration in plugin.json. Trigger when user says "fix hook error", "hook load failed", "plugin hooks broken", "doctor shows hook error", or when /doctor output contains "Hook load failed" or "invalid_type" at path ["hooks"].
What this skill does
# Fix Plugin Hook Loading Errors
Systematic workflow for diagnosing and fixing Claude Code plugin hook validation failures.
## When This Fires
`/doctor` reports something like:
```
Plugin Errors
└ resilient-flows@A2X [resilient-flows]: Hook load failed: [
{ "expected": "record", "code": "invalid_type", "path": ["hooks"], "message": "..." }
]
```
## Step 1 — Read the Error
Parse the `/doctor` output for:
- **Plugin name** and marketplace (e.g., `resilient-flows@A2X`)
- **Zod path** (e.g., `["hooks"]`) — where in the schema validation failed
- **Expected vs received** (e.g., `expected record, received array`)
Common error signatures:
| Error | Meaning |
|-------|---------|
| `expected record, received array` at `["hooks"]` | `hooks` value is `[...]` instead of `{...}` |
| `expected record, received undefined` at `["hooks"]` | Missing `"hooks"` wrapper key (flat format used instead of plugin format) |
| `Duplicate hooks file detected` | `plugin.json` explicitly declares `"hooks": "./hooks/hooks.json"` but the file is also auto-discovered |
## Step 2 — Locate All Copies
**Critical**: The plugin loader reads from CACHED copies, not the source. Find all copies:
```bash
find ~/.claude/plugins -name "hooks.json" -path "*<plugin-name>*" 2>/dev/null
find ~/.claude-work/plugins -name "hooks.json" -path "*<plugin-name>*" 2>/dev/null
```
Typical locations (all must be updated):
1. **Source**: `~/.claude/plugins/marketplaces/<marketplace>/plugins/<plugin>/hooks/hooks.json`
2. **Cache**: `~/.claude-work/plugins/cache/<marketplace>/<plugin>/<version>/hooks/hooks.json`
3. **Marketplaces mirror**: `~/.claude-work/plugins/marketplaces/<marketplace>/plugins/<plugin>/hooks/hooks.json`
## Step 3 — Fix the Schema
The canonical plugin `hooks/hooks.json` format is:
```json
{
"description": "Optional description of what these hooks do",
"hooks": {
"<EventName>": [
{
"matcher": "*",
"hooks": [
{
"type": "command",
"command": "${CLAUDE_PLUGIN_ROOT}/scripts/my-hook.sh",
"timeout": 60
}
]
}
]
}
}
```
**Rules:**
- Top-level `"hooks"` key is **required** and must be a **record/object** (not array)
- `"description"` at top level is **optional**
- Each event name key maps to an **array of matcher groups**
- Each matcher group has `"matcher"` (regex string, `"*"`, or omitted) and `"hooks"` (array of handlers)
- Handler `"type"` must be one of: `command`, `http`, `prompt`, `agent`
- Always use `${CLAUDE_PLUGIN_ROOT}` for portable paths, quoted for spaces
**Validate event names** — only these are valid:
`SessionStart`, `SessionEnd`, `UserPromptSubmit`, `PreToolUse`, `PermissionRequest`, `PermissionDenied`, `PostToolUse`, `PostToolUseFailure`, `Notification`, `SubagentStart`, `SubagentStop`, `TaskCreated`, `TaskCompleted`, `Stop`, `StopFailure`, `TeammateIdle`, `InstructionsLoaded`, `ConfigChange`, `CwdChanged`, `FileChanged`, `WorktreeCreate`, `WorktreeRemove`, `PreCompact`, `PostCompact`, `Elicitation`, `ElicitationResult`
**Note:** `SessionStart`, `CwdChanged`, and `FileChanged` only support `command` hooks (not `prompt`/`agent`/`http`).
For the full hook schema reference, invoke the `hooks-authoring-reference` skill from this plugin.
## Step 4 — Check plugin.json
Read the plugin's `.claude-plugin/plugin.json` and verify:
1. **No explicit hooks declaration** — `hooks/hooks.json` is auto-discovered. If you see `"hooks": "./hooks/hooks.json"`, **remove it**. This causes the "Duplicate hooks file" error.
2. Check all cached copies of plugin.json too (same paths as Step 2 but for `.claude-plugin/plugin.json`).
## Step 5 — Apply Fix to ALL Copies
Write the corrected `hooks.json` to every location found in Step 2. Use the Write tool to update each file. Do NOT skip any cached copy — the loader may read from any of them.
Similarly, if plugin.json needed changes, update all copies.
## Step 6 — Verify
1. Run `/reload-plugins` — check for error count
2. Run `/doctor` — confirm the hook error is gone
3. Run `/hooks` — confirm the plugin's hooks appear under the correct events
4. Optionally: `claude --debug` to see hook registration details
## Common Pitfalls
| Pitfall | Fix |
|---------|-----|
| Fixed source but not cache | Always `find` for ALL copies and update each one |
| Used flat format (events at top level) | Plugin format requires `{"hooks": {...}}` wrapper |
| Used array format (`"hooks": [...]`) | Must be record: `"hooks": {"EventName": [...]}` |
| Added `"hooks"` to plugin.json | Remove it — auto-discovery handles standard path |
| Invalid event name | Check against the 26 valid event names above |
| `prompt`/`agent` hook on SessionStart | SessionStart only supports `command` hooks |
| Missing `"matcher"` field | Optional but recommended — use `"*"` for all |
## Escalation
If the error persists after all fixes:
- Check if the marketplace has a newer version of the plugin
- Try disabling and re-enabling the plugin via `/plugin`
- Clear the plugin cache entirely: `rm -rf ~/.claude-work/plugins/cache/<marketplace>/<plugin>/`
- Then `/reload-plugins` to force a fresh cache build
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.