{
  "markdown": "# ix-claude-plugin\n\nA Claude Code plugin that turns Claude into a **graph-reasoning engineering agent** using [Ix Memory](https://github.com/ix-infrastructure/Ix) as its structured memory backend.\n\nClaude + Ix = reasoning engine + persistent code knowledge graph. Skills are cognitive abstractions (not CLI wrappers) that minimize token usage and maximize accuracy.\n\n## Installation\n\n```\n/plugin marketplace add ix-infrastructure/ix-claude-plugin\n/plugin install ix-memory\n```\n\nRestart Claude Code after installing.\n\nAfter install, Claude is taught to prefer Ix for structural questions, hooks automatically front-run common search/edit paths with Ix context, and Claude appends a short final `Ix` section when hooks materially helped on that turn.\n\n## Requirements\n\n- [Ix Memory](https://github.com/ix-infrastructure/Ix) installed and running (`ix status` returns ok)\n- `jq` in PATH\n- `ripgrep` (`rg`) in PATH\n\n```bash\n# Ubuntu/Debian\nsudo apt install jq ripgrep\n\n# macOS\nbrew install jq ripgrep\n```\n\n**Ix Pro** is optional. All skills and hooks work with basic Ix. Pro adds session briefing injection (goals, bugs, decisions) via the `ix-briefing.sh` hook.\n\n## Skills\n\nHigh-level cognitive skills — each one infers intent, orchestrates multiple graph queries, and synthesizes output. None are CLI aliases.\n\n| Skill | What it does | Key rule |\n|-------|-------------|----------|\n| `/ix-understand [target]` | Build a mental model of a system or the whole repo | Graph only — no source reads |\n| `/ix-investigate <symbol>` | Deep dive: what it is, how it connects, execution path | Graph first, one symbol read max |\n| `/ix-impact <target>` | Change risk: blast radius, affected systems, test targets | Depth scales with risk level |\n| `/ix-plan <targets...>` | Risk-ordered implementation plan for a set of changes | Parallel impact, finds shared dependents |\n| `/ix-debug <symptom>` | Root cause analysis from symptom to candidates | Targeted reads at suspects only |\n| `/ix-architecture [scope]` | Design health: coupling, smells, hotspots | Graph only — never reads source |\n| `/ix-docs <target> [--full] [--style narrative|reference|hybrid] [--split] [--single-doc] [--out <path>]` | Generate narrative-first system documentation with a selective reference layer | Default is onboarding-focused; `--full --style hybrid` gives the deepest coverage |\n\nAll skills fall back gracefully when ix is unavailable.\n\n## Agents\n\nAutonomous multi-step agents for complex tasks:\n\n| Agent | Purpose |\n|-------|---------|\n| `ix-explorer` | General-purpose graph exploration, open-ended questions |\n| `ix-system-explorer` | Full architectural model of a codebase or region |\n| `ix-bug-investigator` | Autonomous investigation from symptom to root cause candidates |\n| `ix-safe-refactor-planner` | Blast radius + safe change sequencing for refactors |\n| `ix-architecture-auditor` | Full structural health report with ranked improvements |\n\n## Automatic hooks\n\n| Trigger | Hook | Effect |\n|---------|------|--------|\n| User sends any prompt | `UserPromptSubmit` → `ix-briefing.sh` | Injects session briefing (goals, bugs, decisions) once per 10 min — **requires Ix Pro**; also instructs Claude to append a short final `Ix` section when hooks helped |\n| Claude runs `Grep` or `Glob` | `PreToolUse` → `ix-intercept.sh` | Front-runs with `ix text` + `ix locate`/`ix inventory` |\n| Claude runs `Bash` with grep/rg | `PreToolUse` → `ix-bash.sh` | Extracts pattern, front-runs with `ix text` + `ix locate` |\n| Claude edits a file | `PreToolUse` → `ix-pre-edit.sh` | Runs `ix impact` before the edit |\n| Claude edits a file | `PostToolUse` → `ix-ingest.sh` (async) | Runs `ix map <file>` to update the graph |\n| Claude finishes responding | `Stop` → `ix-annotate.sh` | Emits a factual summary of how ix helped on that turn on non-`modelSuffix` channels |\n| Claude finishes responding | `Stop` → `ix-map.sh` (async) | Runs `ix map` to refresh the full graph |\n\n`hooks/ix-read.sh` remains in the repo as a disabled placeholder, but it is not registered in `hooks/hooks.json` and does not run at runtime.\n\nAll hooks bail silently if `ix` is not in PATH or the backend is unreachable.\n\n## Debugging ix Hook Calls\n\nIf you want to see exactly which `ix` CLI commands Claude-triggered hooks run, start Claude Code with `IX_DEBUG_LOG` set in its environment:\n\n```bash\nexport IX_DEBUG_LOG=/tmp/ix-hooks.log\ntail -f /tmp/ix-hooks.log\n```\n\nThe hooks will append lines like:\n\n```text\n[2026-04-19T12:34:56-0700] [ix-intercept] CMD ix text AuthService --limit 15 --format json --path src/\n[2026-04-19T12:34:56-0700] [ix-intercept] CMD ix locate AuthService --format json\n```\n\n`CMD ...` lines are the exact shell-escaped `ix` invocations. `INJECT ...` lines show the exact context a hook injected back into Claude.\n\n## Uninstall\n\n```\n/plugin uninstall ix-memory\n```\n",
  "bytes": 4831,
  "sha": "d9fc64a6282a4d760d26e0bc423f8cb49d619c9b08f65023ff0011833610962f",
  "repo_slug": "ix-infrastructure/ix-claude-plugin",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_ix_infrastructure_ix_claude_plugin_ix_me_32eeb71a/readme"
}