agentsys 5.6.4 → 5.7.0

package/README.md

@@ -19,7 +19,7 @@
 </p>
 
 <p align="center">
-  <b>19 plugins · 38 agents · 39 skills (across all repos) · 30k lines of lib code · 3,575 tests · 5 platforms</b><br>
+  <b>19 plugins · 47 agents · 39 skills (across all repos) · 30k lines of lib code · 3,583 tests · 5 platforms</b><br>
   <em>Plugins distributed as standalone repos under <a href="https://github.com/agent-sh">agent-sh</a> org — agentsys is the marketplace &amp; installer</em>
 </p>
 
@@ -45,7 +45,7 @@ AI models can write code. That's not the hard part anymore. The hard part is eve
 
 ## What This Is
 
-An agent orchestration system — 19 plugins, 38 agents, and 39 skills that compose into structured pipelines for software development. Each plugin lives in its own standalone repo under the [agent-sh](https://github.com/agent-sh) org. agentsys is the marketplace and installer that ties them together.
+An agent orchestration system — 19 plugins, 47 agents, and 39 skills that compose into structured pipelines for software development. Each plugin lives in its own standalone repo under the [agent-sh](https://github.com/agent-sh) org. agentsys is the marketplace and installer that ties them together.
 
 Each agent has a single responsibility, a specific model assignment, and defined inputs/outputs. Pipelines enforce phase gates so agents can't skip steps. State persists across sessions so work survives interruptions.
 
@@ -73,9 +73,46 @@ This came from testing on 1,000+ repositories.
 
 ---
 
+## Benchmarks
+
+Structured prompts and enriched context do more for output quality than model tier. Benchmarked March 2026 on real tasks (`/can-i-help` and `/onboard` against [glide-mq](https://github.com/avifenesh/glide-mq)), measured with `claude -p --output-format json`. Models: Claude Opus 4 and Claude Sonnet 4.
+
+### Sonnet + AgentSys vs raw Opus
+
+Same task, same repo, same prompt ("I want to improve docs"):
+
+| Configuration | Cost | Output tokens | Result quality |
+|---------------|------|---------------|----------------|
+| Opus, no agentsys | $1.10 | 2,841 | Generic recommendations, no project-specific context |
+| Opus + agentsys | $1.95 | 5,879 | Specific recommendations with effort estimates, convention awareness, breaking change detection |
+| **Sonnet + agentsys** | **$0.66** | **6,084** | **Comparable to Opus + agentsys: specific, actionable, project-aware** |
+
+Sonnet + agentsys produced more output with higher specificity than raw Opus - at 40% lower cost.
+
+### With agentsys, model tier matters less
+
+Once the pipeline provides structured prompts, enriched repo-intel data, and phase-gated workflows, the model does less heavy lifting. The gap between Sonnet and Opus narrows:
+
+| Plugin | Opus | Sonnet | Savings |
+|--------|------|--------|---------|
+| /onboard | $1.10 | $0.30 | 73% |
+| /can-i-help | $1.34 | $0.23 | 83% |
+
+Both models reached the same outcome quality - Sonnet just costs less to get there. The structured pipeline captures most of the gains that would otherwise require a more expensive model.
+
+### What this means
+
+| Scenario | Model cost | Quality |
+|----------|-----------|---------|
+| Without agentsys | Need Opus for good results | Depends on model capability |
+| **With agentsys** | **Sonnet is sufficient** | **Pipeline handles the structure, model handles judgment** |
+
+The investment shifts from model spend to pipeline design. Better prompts, richer context, enforced phases - these compound in ways that model upgrades alone don't.
+
+---
+
 ## Commands
 
-<!-- GEN:START:readme-commands -->
 | Command | What it does |
 |---------|--------------|
 | [`/next-task`](#next-task) | Task workflow: discovery, implementation, PR, merge |
@@ -86,7 +123,7 @@ This came from testing on 1,000+ repositories.
 | [`/drift-detect`](#drift-detect) | Compare plan vs implementation |
 | [`/audit-project`](#audit-project) | Multi-agent iterative code review |
 | [`/enhance`](#enhance) | Plugin, agent, and prompt analyzers |
-| [`/repo-map`](#repo-map) | AST-based repository map |
+| [`/repo-intel`](#repo-intel) | Unified static analysis - git history, AST symbols, project metadata |
 | [`/sync-docs`](#sync-docs) | Sync documentation with code changes |
 | [`/learn`](#learn) | Research topics, create learning guides |
 | [`/consult`](#consult) | Cross-tool AI consultation |
@@ -94,10 +131,8 @@ This came from testing on 1,000+ repositories.
 | [`/web-ctl`](#web-ctl) | Browser automation for AI agents |
 | [`/release`](#release) | Versioned release with ecosystem detection |
 | [`/skillers`](#skillers) | Workflow pattern learning and automation |
-| [`/git-map`](#git-map) | Git history analysis: hotspots, coupling, ownership, bus factor |
 | [`/onboard`](#onboard) | Codebase orientation for newcomers |
 | [`/can-i-help`](#can-i-help) | Match contributor skills to project needs |
-<!-- GEN:END:readme-commands -->
 
 Each command works standalone. Together, they compose into end-to-end pipelines.
 
@@ -105,12 +140,12 @@ Each command works standalone. Together, they compose into end-to-end pipelines.
 
 ## Skills
 
-<!-- GEN:START:readme-skills -->
-39 skills included across the plugins:
+38 skills included across the plugins:
 
 | Category | Skills |
 |----------|--------|
 | **Workflow** | `discover-tasks`, `orchestrate-review`, `validate-delivery` |
+| **Message Queues** | `glide-mq-migrate-bee`, `glide-mq-migrate-bullmq`, `glide-mq` |
 | **Enhancement** | `enhance-agent-prompts`, `enhance-claude-memory`, `enhance-cross-file`, `enhance-docs`, `enhance-hooks`, `enhance-orchestrator`, `enhance-plugins`, `enhance-prompts`, `enhance-skills` |
 | **Performance** | `baseline`, `benchmark`, `code-paths`, `investigation-logger`, `perf-analyzer`, `profile`, `theory-gatherer`, `theory-tester` |
 | **Cleanup** | `deslop`, `sync-docs` |
@@ -119,9 +154,7 @@ Each command works standalone. Together, they compose into end-to-end pipelines.
 | **Onboarding** | `can-i-help`, `onboard` |
 | **Web** | `web-auth`, `web-browse` |
 | **Release** | `release` |
-| **Analysis** | `drift-analysis`, `git-mapping`, `repo-mapping` |
-| **Other** | `glide-mq-migrate-bee`, `glide-mq-migrate-bullmq`, `glide-mq` |
-<!-- GEN:END:readme-skills -->
+| **Analysis** | `drift-analysis`, `repo-intel` |
 
 **External skill plugins** (standalone repos, installed separately):
 
@@ -138,8 +171,10 @@ Skills are the reusable implementation units. Agents invoke skills; commands orc
 | Section | What's there |
 |---------|--------------|
 | [The Approach](#the-approach) | Why it's built this way |
+| [Benchmarks](#benchmarks) | Sonnet + agentsys vs raw Opus |
 | [Commands](#commands) | All 19 commands overview |
 | [Skills](#skills) | 39 skills across plugins |
+| [Skill-Only Plugins](#skill-only-plugins) | glide-mq and other non-command plugins |
 | [Command Details](#command-details) | Deep dive into each command |
 | [How Commands Work Together](#how-commands-work-together) | Standalone vs integrated |
 | [Design Philosophy](#design-philosophy) | The thinking behind the architecture |
@@ -149,6 +184,26 @@ Skills are the reusable implementation units. Agents invoke skills; commands orc
 
 ---
 
+## Skill-Only Plugins
+
+Plugins that provide skills without a `/` command. Installed alongside agentsys; skills become available to all agents.
+
+### glide-mq
+
+Build message queues, background jobs, and workflow orchestration with [glide-mq](https://github.com/avifenesh/glide-mq) - high-performance Node.js queue on Valkey/Redis.
+
+| Skill | What it does |
+|-------|--------------|
+| `glide-mq` | Greenfield queue development - queues, workers, ordering, rate limiting, flows, broadcast, step jobs |
+| `glide-mq-migrate-bullmq` | Migrate from BullMQ to glide-mq - API mapping, breaking changes, feature comparison |
+| `glide-mq-migrate-bee` | Migrate from Bee-Queue to glide-mq - API mapping, pattern conversion |
+
+Key features: per-key ordering, group concurrency, runtime group rate limiting (`job.rateLimitGroup()`), token bucket, DAG workflows, broadcast pub/sub, step jobs, deduplication, serverless producers.
+
+[Skill plugin →](https://github.com/agent-sh/glidemq) | [glide-mq docs →](https://avifenesh.github.io/glide-mq.dev/) | [npm →](https://www.npmjs.com/package/glide-mq)
+
+---
+
 ## Command Details
 
 ### /next-task
@@ -399,7 +454,7 @@ Three phases run in sequence:
 3. **Breaking Point** - Binary search to find failure threshold
 4. **Constraints** - CPU/memory limits, measure delta vs baseline
 5. **Hypotheses** - Generate up to 5 hypotheses with evidence and confidence
-6. **Code Paths** - Use repo-map to identify entrypoints and hot files
+6. **Code Paths** - Use repo-intel to identify entrypoints and hot files
 7. **Profiling** - Language-specific tools (--cpu-prof, JFR, cProfile, pprof)
 8. **Optimization** - One change per experiment, 2+ validation passes
 9. **Decision** - Continue or stop based on measurable improvement
@@ -546,30 +601,33 @@ Findings are collected and categorized by severity (critical/high/medium/low). A
 
 ---
 
-### /repo-map
+### /repo-intel
 
-**Purpose:** Builds an AST-based map of symbols and imports for fast repo analysis.
+**Purpose:** Unified static analysis - git history, AST symbols, and project metadata in one plugin.
 
-**What it generates:**
+**What it provides:**
 
-- Cached file→symbols map (exports, functions, classes)
-- Import graph for dependency hints
+- Git history intelligence: hotspots, coupling, ownership, bus factor, bugspots, AI detection
+- AST symbol mapping: exports, functions, classes, imports
+- Project metadata and health metrics
 
-Output is cached at `{state-dir}/repo-map.json` and exposed via the MCP `repo_map` tool.
+Output is cached at `{state-dir}/repo-intel.json` and `{state-dir}/repo-map.json`.
 
 **Why it matters:**
 
-Tools like `/drift-detect` and planners can use the map instead of re-scanning the repo every time.
+Tools like `/drift-detect`, `/onboard`, `/can-i-help`, and planners consume this data instead of re-scanning the repo every time. 9 plugins use repo-intel data automatically.
 
 **Usage:**
 
 ```bash
-/repo-map init        # First-time map generation
-/repo-map update      # Incremental update
-/repo-map status      # Check freshness
+/repo-intel init                   # First-time scan
+/repo-intel update                 # Incremental update
+/repo-intel query hotspots         # Most active files
+/repo-intel query ownership src/   # Who owns a path
+/repo-intel query bus-factor       # Knowledge risk
 ```
 
-**Required:** ast-grep (`sg`) must be installed.
+Backed by [agent-analyzer](https://github.com/agent-sh/agent-analyzer) Rust binary.
 
 ---
 
@@ -848,44 +906,6 @@ No per-turn overhead - it reads transcripts that Claude Code already saves.
 
 ---
 
-### /git-map
-
-**Purpose:** Analyze git history to surface hotspots, coupling, ownership, bus factor, bugspots, area health, and AI attribution.
-
-**How it works:**
-
-The plugin wraps the [agent-analyzer](https://github.com/agent-sh/agent-analyzer) Rust binary. Run `init` once to scan git history and cache the result as `repo-intel.json`. Then run queries instantly.
-
-**20 query types:**
-
-| Category | Queries |
-|----------|---------|
-| Activity | `hotspots`, `coldspots`, `file-history` |
-| Quality | `bugspots`, `test-gaps`, `diff-risk` |
-| People | `ownership`, `contributors`, `bus-factor` |
-| Coupling | `coupling` |
-| Standards | `norms`, `conventions` |
-| Health | `areas`, `health`, `release-info` |
-| AI | `ai-ratio`, `recent-ai` |
-| Guidance | `onboard`, `can-i-help` |
-| Docs | `doc-drift` |
-
-**9 plugins consume git-map data automatically** - deslop, sync-docs, drift-detect, audit-project, next-task, enhance, ship, onboard, can-i-help.
-
-**Usage:**
-
-```bash
-/git-map init                      # First-time scan
-/git-map update                    # Add new commits
-/git-map query hotspots            # Most active files
-/git-map query ownership src/      # Who owns a path
-/git-map query bus-factor          # Knowledge risk
-```
-
-[Full query reference →](https://github.com/agent-sh/git-map)
-
----
-
 ### /onboard
 
 **Purpose:** Get oriented in any codebase in under 3 minutes.
@@ -904,7 +924,7 @@ The plugin wraps the [agent-analyzer](https://github.com/agent-sh/agent-analyzer
 |-------|------|------|
 | quick | ~2s | Manifest + README + structure |
 | normal | ~5s | + CLAUDE.md/AGENTS.md + CI + repo-intel |
-| deep | ~15s | + repo-map AST symbols |
+| deep | ~15s | + repo-intel AST symbols |
 
 **Supported manifests:** package.json, Cargo.toml, go.mod, pyproject.toml, deno.json, CMakeLists.txt, meson.build, setup.py, pom.xml, build.gradle. Detects monorepos (npm/pnpm/lerna/Cargo workspaces, Python libs/, Deno workspaces).
 
@@ -956,28 +976,6 @@ The plugin wraps the [agent-analyzer](https://github.com/agent-sh/agent-analyzer
 
 ---
 
-## Skill-Only Plugins
-
-Plugins that provide skills without a `/` command. Installed alongside agentsys; skills become available to all agents.
-
-### glide-mq
-
-**Purpose:** Build message queues, background jobs, and workflow orchestration with [glide-mq](https://github.com/avifenesh/glide-mq) - high-performance Node.js queue on Valkey/Redis.
-
-**Skills:**
-
-| Skill | What it does |
-|-------|--------------|
-| `glide-mq` | Greenfield queue development - queues, workers, ordering, rate limiting, flows, broadcast, step jobs |
-| `glide-mq-migrate-bullmq` | Migrate from BullMQ to glide-mq - API mapping, breaking changes, feature comparison |
-| `glide-mq-migrate-bee` | Migrate from Bee-Queue to glide-mq - API mapping, pattern conversion |
-
-**Key features covered:** per-key ordering, group concurrency, runtime group rate limiting (`job.rateLimitGroup()`), token bucket, DAG workflows, broadcast pub/sub, step jobs, deduplication, serverless producers.
-
-[Full documentation →](https://github.com/agent-sh/glidemq) | [glide-mq docs →](https://avifenesh.github.io/glide-mq.dev/)
-
----
-
 ## How Commands Work Together
 
 **Standalone use:**
@@ -1154,8 +1152,8 @@ agentsys --development              # Dev mode (bypasses marketplace)
 **For GitLab workflows:**
 - GitLab CLI (`glab`) authenticated
 
-**For /repo-map:**
-- ast-grep (`sg`) installed
+**For /repo-intel:**
+- agent-analyzer (installed automatically via npm)
 
 **For /agnix:**
 - [agnix CLI](https://github.com/agent-sh/agnix) installed (`npm install -g agnix`, `cargo install agnix-cli`, or `brew install agnix`)
@@ -1181,7 +1179,7 @@ The system is built on research, not guesswork.
 - Instruction following reliability
 
 **Testing:**
-- 3,575 tests passing
+- 3,583 tests passing
 - Drift-detect validated on 1,000+ repositories
 - E2E workflow testing across all commands
 - Cross-platform validation (Claude Code, OpenCode, Codex CLI, Cursor, Kiro)

package/lib/binary/version.js

@@ -1,7 +1,7 @@
 'use strict';
 
 // Minimum binary version required by this version of agent-core
-const ANALYZER_MIN_VERSION = '0.1.0';
+const ANALYZER_MIN_VERSION = '0.3.0';
 
 // Binary name
 const BINARY_NAME = 'agent-analyzer';

package/lib/repo-map/converter.js

@@ -0,0 +1,130 @@
+'use strict';
+
+/**
+ * Convert agent-analyzer repo-intel.json format to repo-map.json format.
+ *
+ * agent-analyzer outputs: { symbols: { [filePath]: { exports, imports, definitions } } }
+ * repo-map expects:       { files: { [filePath]: { language, symbols, imports } } }
+ *
+ * @module lib/repo-map/converter
+ */
+
+const path = require('path');
+
+const LANGUAGE_BY_EXTENSION = {
+  '.js': 'javascript', '.jsx': 'javascript', '.mjs': 'javascript', '.cjs': 'javascript',
+  '.ts': 'typescript', '.tsx': 'typescript', '.mts': 'typescript', '.cts': 'typescript',
+  '.py': 'python', '.pyw': 'python',
+  '.rs': 'rust',
+  '.go': 'go',
+  '.java': 'java'
+};
+
+// SymbolKind values from agent-analyzer (kebab-case serialized)
+const CLASS_KINDS = new Set(['class', 'struct', 'interface', 'enum', 'impl']);
+const TYPE_KINDS = new Set(['trait', 'type-alias']);
+const FUNCTION_LIKE_KINDS = new Set(['method', 'arrow', 'closure']);
+const CONSTANT_KINDS = new Set(['constant', 'variable', 'const', 'field', 'property']);
+
+function detectLanguage(filePath) {
+  return LANGUAGE_BY_EXTENSION[path.extname(filePath).toLowerCase()] || 'unknown';
+}
+
+function detectLanguagesFromFiles(filePaths) {
+  const langs = new Set();
+  for (const fp of filePaths) {
+    const lang = detectLanguage(fp);
+    if (lang !== 'unknown') langs.add(lang);
+  }
+  return Array.from(langs);
+}
+
+/**
+ * Convert a single file's symbols from repo-intel format to repo-map format.
+ * @param {string} filePath
+ * @param {Object} fileSym - { exports, imports, definitions }
+ * @returns {Object} repo-map file entry
+ */
+function convertFile(filePath, fileSym) {
+  const exportNames = new Set((fileSym.exports || []).map(e => e.name));
+
+  const exports = (fileSym.exports || []).map(e => ({
+    name: e.name,
+    kind: e.kind,
+    line: e.line
+  }));
+
+  const functions = [];
+  const classes = [];
+  const types = [];
+  const constants = [];
+
+  for (const def of fileSym.definitions || []) {
+    const entry = {
+      name: def.name,
+      kind: def.kind,
+      line: def.line,
+      exported: exportNames.has(def.name)
+    };
+    if (def.kind === 'function' || FUNCTION_LIKE_KINDS.has(def.kind)) {
+      functions.push(entry);
+    } else if (CLASS_KINDS.has(def.kind)) {
+      classes.push(entry);
+    } else if (TYPE_KINDS.has(def.kind)) {
+      types.push(entry);
+    } else if (CONSTANT_KINDS.has(def.kind)) {
+      constants.push(entry);
+    } else {
+      // Unknown kind - default to constants for backward compat
+      constants.push(entry);
+    }
+  }
+
+  // agent-analyzer imports: [{ from, names }] → repo-map imports: [{ source, kind, names }]
+  const imports = (fileSym.imports || []).map(imp => ({
+    source: imp.from,
+    kind: 'import',
+    names: imp.names || []
+  }));
+
+  return {
+    language: detectLanguage(filePath),
+    symbols: { exports, functions, classes, types, constants },
+    imports
+  };
+}
+
+/**
+ * Convert a full repo-intel data object to repo-map format.
+ * @param {Object} intel - RepoIntelData from agent-analyzer
+ * @returns {Object} repo-map.json compatible object
+ */
+function convertIntelToRepoMap(intel) {
+  const files = {};
+  let totalSymbols = 0;
+  let totalImports = 0;
+
+  for (const [filePath, fileSym] of Object.entries(intel.symbols || {})) {
+    files[filePath] = convertFile(filePath, fileSym);
+    const s = files[filePath].symbols;
+    totalSymbols += s.functions.length + s.classes.length +
+                    s.types.length + s.constants.length;
+    totalImports += files[filePath].imports.length;
+  }
+
+  return {
+    version: '2.0',
+    generated: intel.generated || new Date().toISOString(),
+    git: intel.git ? { commit: intel.git.analyzedUpTo } : undefined,
+    project: { languages: detectLanguagesFromFiles(Object.keys(files)) },
+    stats: {
+      totalFiles: Object.keys(files).length,
+      totalSymbols,
+      totalImports,
+      errors: []
+    },
+    files
+  };
+}
+
+module.exports = { convertIntelToRepoMap, convertFile, detectLanguage };