agentsys 5.11.0 → 5.13.0

package/.claude-plugin/marketplace.json

@@ -1,7 +1,7 @@
 {
   "name": "agentsys",
-  "description": "20 specialized plugins for AI workflow automation - task orchestration, PR workflow, slop detection, code review, drift detection, enhancement analysis, documentation sync, unified static analysis, perf investigations, topic research, agent config linting, cross-tool AI consultation, structured AI debate, workflow pattern learning, codebase onboarding, contributor guidance, and Zig language support",
-  "version": "5.11.0",
+  "description": "24 specialized plugins for AI workflow automation - task orchestration, PR workflow, slop detection, code review, drift detection, enhancement analysis, documentation sync, unified static analysis, durable memory, negative behavior memory, skill and system prompt curation, perf investigations, topic research, agent config linting, cross-tool AI consultation, structured AI debate, workflow pattern learning, codebase onboarding, contributor guidance, and Zig language support",
+  "version": "5.13.0",
   "owner": {
     "name": "Avi Fenesh",
     "url": "https://github.com/avifenesh"
@@ -40,11 +40,11 @@
       "source": {
         "source": "url",
         "url": "https://github.com/agent-sh/prepare-delivery.git",
-        "commit": "693d7649501608da49aea8efe610a7029100b8f2",
-        "ref": "v0.1.1"
+        "commit": "0e9685fe0e93e058af5ca9a0374f4f97e1db878c",
+        "ref": "v0.1.2"
       },
       "description": "Pre-ship quality gates: deslop, simplify, agnix, enhance, review loop, delivery validation, docs sync",
-      "version": "0.1.1",
+      "version": "0.1.2",
       "category": "productivity",
       "homepage": "https://github.com/agent-sh/prepare-delivery"
     },
@@ -78,7 +78,7 @@
       "source": {
         "source": "url",
         "url": "https://github.com/agent-sh/deslop.git",
-        "commit": "dc49a5309a104a011439f87b346d1c3b47375db2"
+        "commit": "00301b9ce81d12caa38063e4a65b535ba5b011b2"
       },
       "description": "3-phase AI slop detection: regex patterns (HIGH), multi-pass analyzers (MEDIUM), CLI tools (LOW)",
       "version": "1.0.0",
@@ -90,11 +90,11 @@
       "source": {
         "source": "url",
         "url": "https://github.com/agent-sh/audit-project.git",
-        "commit": "2c961a84ab5945670be44d6c54eb496099effe48",
-        "ref": "v1.0.1"
+        "commit": "f703facec38765b6fd8cb5a2076c98bf14e5998e",
+        "ref": "v1.0.2"
       },
       "description": "Multi-agent iterative code review until zero issues remain",
-      "version": "1.0.1",
+      "version": "1.0.2",
       "category": "development",
       "homepage": "https://github.com/agent-sh/audit-project"
     },
@@ -103,7 +103,7 @@
       "source": {
         "source": "url",
         "url": "https://github.com/agent-sh/drift-detect.git",
-        "commit": "e7edd602a0e24ae2bff8e262c58e481aae448944"
+        "commit": "576aca37402068aef175c8a6002584e19cb6ea74"
       },
       "description": "Deep repository analysis to realign project plans with code reality - detects drift, gaps, and creates prioritized reconstruction plans",
       "version": "1.0.0",
@@ -115,19 +115,45 @@
       "source": {
         "source": "url",
         "url": "https://github.com/agent-sh/enhance.git",
-        "commit": "31e7d3861afb20b3d910be957387de23dc3ae854"
+        "commit": "93f299e494a3a9ea74e82bd2d15bc1c517ce8f0c"
       },
       "description": "Master enhancement orchestrator: parallel analyzer execution for plugins, agents, docs, CLAUDE.md, and prompts with unified reporting",
       "version": "1.0.0",
       "category": "development",
       "homepage": "https://github.com/agent-sh/enhance"
     },
+    {
+      "name": "skill-curator",
+      "source": {
+        "source": "url",
+        "url": "https://github.com/agent-sh/skill-curator.git",
+        "ref": "v1.0.0",
+        "commit": "e30ebfe3e7a5307e97fa4e6d59f538b12f8074c5"
+      },
+      "description": "Canonical guidance for writing, reviewing, and maintaining production-grade SKILL.md files across agent tools",
+      "version": "1.0.0",
+      "category": "development",
+      "homepage": "https://github.com/agent-sh/skill-curator"
+    },
+    {
+      "name": "system-prompt-curator",
+      "source": {
+        "source": "url",
+        "url": "https://github.com/agent-sh/system-prompt-curator.git",
+        "ref": "v2.0.0",
+        "commit": "277bee4b8b3b796d2a627132c88fb132479cbde0"
+      },
+      "description": "Production-grade system prompt curation for autonomous coding agents and multi-agent orchestration",
+      "version": "2.0.0",
+      "category": "development",
+      "homepage": "https://github.com/agent-sh/system-prompt-curator"
+    },
     {
       "name": "sync-docs",
       "source": {
         "source": "url",
         "url": "https://github.com/agent-sh/sync-docs.git",
-        "commit": "961f5e67b583f70a4e0ad3ad83503ab90decc11d"
+        "commit": "410e06739da101583b7238669f4acad7f5aea7ab"
       },
       "description": "Standalone documentation sync: find outdated refs, update CHANGELOG, flag stale examples based on code changes",
       "version": "1.0.0",
@@ -165,13 +191,38 @@
       "source": {
         "source": "url",
         "url": "https://github.com/agent-sh/learn.git",
-        "commit": "e28ea11f4622509d1ae7425fa97f7dd719d43716"
+        "commit": "b3025d376a83841078f3ab5cf53b62c6960e46c3"
       },
       "description": "Research topics online and create comprehensive learning guides with RAG-optimized indexes",
       "version": "1.0.0",
       "category": "productivity",
       "homepage": "https://github.com/agent-sh/learn"
     },
+    {
+      "name": "axiom",
+      "source": {
+        "source": "url",
+        "url": "https://github.com/agent-sh/axiom.git",
+        "commit": "a1bf0fc9ab61178774b3391a337db3e1af20413a"
+      },
+      "description": "Personal agent-native knowledge base: load thin context, query durable memories, create project scaffolds, and propose approved records",
+      "version": "0.6.0",
+      "category": "productivity",
+      "homepage": "https://github.com/agent-sh/axiom"
+    },
+    {
+      "name": "banthis",
+      "source": {
+        "source": "url",
+        "url": "https://github.com/agent-sh/banthis.git",
+        "ref": "v0.3.0",
+        "commit": "5820a197bba0f01ec4877a63c5d13c108d8c661d"
+      },
+      "description": "Durable negative behavior memory: persist user-defined banned agent behaviors into CLAUDE.md or AGENTS.md",
+      "version": "0.3.0",
+      "category": "productivity",
+      "homepage": "https://github.com/agent-sh/banthis"
+    },
     {
       "name": "agnix",
       "source": {
@@ -190,7 +241,7 @@
       "source": {
         "source": "url",
         "url": "https://github.com/agent-sh/consult.git",
-        "commit": "3115688a7c6079a04e8caa0183c1ca020e3d413d"
+        "commit": "8ce96f86c0ae67f732383a8b45c9994a7cd64d2a"
       },
       "description": "Cross-tool AI consultation: get second opinions from Gemini CLI, Codex CLI, Claude Code, OpenCode, or Copilot CLI with model and thinking effort control",
       "version": "1.0.0",

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "agentsys",
-  "version": "5.11.0",
+  "version": "5.13.0",
   "description": "Professional-grade slash commands for Claude Code with cross-platform support",
   "keywords": [
     "workflow",

package/.codex-plugin/plugin.json

@@ -4,8 +4,8 @@
   "skills": "./adapters/codex/skills",
   "interface": {
     "displayName": "agentsys",
-    "shortDescription": "AI agent orchestration with 42 skills and 50 agents",
-    "longDescription": "Professional-grade slash commands for AI-powered development workflows. Includes /next-task (task discovery to production), /ship (commit to deploy), /audit-project (multi-agent code review), /deslop (AI slop cleanup), /perf (performance investigation), /enhance (config analysis), /consult (cross-tool AI consultation), and more.",
+    "shortDescription": "AI agent orchestration with 45 skills and 49 agents",
+    "longDescription": "Professional-grade slash commands for AI-powered development workflows. Includes /next-task (task discovery to production), /axiom (durable agent memory), /banthis (negative behavior memory), /skill-curator and /system-prompt-curator (agent configuration curation), /ship (commit to deploy), /audit-project (multi-agent code review), /deslop (AI slop cleanup), /perf (performance investigation), /enhance (config analysis), /consult (cross-tool AI consultation), and more.",
     "developerName": "Avi Fenesh",
     "category": "developer-tools",
     "capabilities": [
@@ -14,12 +14,13 @@
       "ci-cd",
       "deployment",
       "performance",
-      "documentation"
+      "documentation",
+      "memory"
     ],
     "websiteUrl": "https://agent-sh.github.io/agent-sh.dev/",
     "defaultPrompt": [
       "What should I work on next?",
-      "Ship my changes",
+      "Load my durable Axiom context",
       "Review this codebase"
     ]
   }

package/AGENTS.md

@@ -76,7 +76,7 @@
 <!-- GEN:START:claude-architecture -->
 ```
 lib/          → Shared library (vendored to plugins)
-plugins/      → 0 plugins, 10 agents (0 file-based + 10 role-based), 0 skills
+plugins/      → 24 plugins, 49 agents (39 file-based + 10 role-based), 45 skills
 adapters/     → Platform adapters (opencode-plugin/, opencode/, codex/)
 checklists/   → Action checklists (9 files)
 bin/cli.js    → npm CLI installer
@@ -84,6 +84,30 @@ bin/cli.js    → npm CLI installer
 
 | Plugin | Agents | Skills | Purpose |
 |--------|--------|--------|---------|
+| next-task | 8 | 1 | Master workflow orchestration |
+| prepare-delivery | 3 | 4 | Pre-ship quality gates |
+| gate-and-ship | 0 | 0 | Quality gates then ship |
+| ship | 1 | 1 | PR creation and deployment |
+| deslop | 1 | 1 | AI slop cleanup |
+| audit-project | 10 | 1 | Multi-agent code review |
+| drift-detect | 1 | 1 | Plan drift detection |
+| enhance | 8 | 9 | Code quality analyzers |
+| sync-docs | 1 | 1 | Documentation sync |
+| repo-intel | 1 | 1 | Unified static analysis |
+| axiom | 0 | 1 | Durable agent-native memory |
+| banthis | 0 | 1 | Durable negative behavior memory |
+| perf | 6 | 8 | Performance investigation |
+| learn | 1 | 1 | Topic research and learning guides |
+| agnix | 1 | 1 | Agent config linting |
+| consult | 1 | 1 | Cross-tool AI consultation |
+| debate | 1 | 1 | Multi-perspective debate analysis |
+| web-ctl | 1 | 2 | Browser automation for AI agents |
+| skill-curator | 0 | 1 | Skill authoring and review |
+| system-prompt-curator | 0 | 1 | System prompt curation |
+| skillers | 2 | 2 | Workflow pattern learning |
+| onboard | 1 | 1 | Codebase onboarding |
+| can-i-help | 1 | 1 | Contributor guidance |
+| zig-lsp | 0 | 0 |  |
 <!-- GEN:END:claude-architecture -->
 
 **Pattern**: `Command → Agent → Skill` (orchestration → invocation → implementation)
@@ -152,7 +176,7 @@ agentsys                # Run installer
 <agents>
 ## Agents
 
-49 agents across 20 plugins (18 have agents; gate-and-ship is commands-only; zig-lsp is config-only with no commands or agents). Key agents by model:
+49 agents across 24 plugins (18 have agents; gate-and-ship is commands-only; axiom, banthis, skill-curator, and system-prompt-curator are skill/command-only; zig-lsp is config-only with no commands or agents). Key agents by model:
 
 | Model | Agents | Use Case |
 |-------|--------|----------|
@@ -166,14 +190,15 @@ See [README.md](./README.md#command-details) and [docs/reference/AGENTS.md](./do
 <skills>
 ## Skills
 
-40 skills across plugins. Agents invoke skills for reusable implementation.
+45 skills across plugins. Agents invoke skills for reusable implementation.
 
 | Category | Key Skills |
 |----------|------------|
 | Workflow | `orchestrate-review`, `discover-tasks`, `validate-delivery` |
-| Enhancement | `enhance-*` (9 skills for plugins, agents, docs, prompts, hooks) |
+| Enhancement | `enhance-*` (9 skills for plugins, agents, docs, prompts, hooks), `skill-curator`, `system-prompt-curator` |
 | Performance | `baseline`, `benchmark`, `profile`, `theory-tester` |
 | Cleanup | `deslop`, `sync-docs`, `drift-analysis`, `repo-intel` |
+| Memory | `axiom`, `banthis` |
 
 See [README.md](./README.md#skills) for full skill list.
 </skills>

package/CHANGELOG.md

@@ -9,6 +9,24 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [5.13.0] - 2026-05-17
+
+### Added
+- Added the `axiom` plugin to the agentsys marketplace and static docs metadata. Axiom provides durable agent-native memory, project context loading, scoped querying, project scaffolding, and human-approved record proposals.
+- Added `banthis` as a published standalone plugin pinned to `v0.3.0`. It provides durable negative behavior memory through a tiny CLI, skill, and slash command.
+- Added `skill-curator` as a published standalone plugin pinned to `v1.0.0` for production-grade `SKILL.md` authoring and review.
+- Added `system-prompt-curator` as a published standalone plugin pinned to `v2.0.0` for autonomous coding-agent system prompt curation.
+
+### Changed
+- Updated marketplace, Codex metadata, docs, and website counts to 24 plugins, 49 agents, and 45 skills.
+
+## [5.12.0] - 2026-04-26
+
+### Propagated upstream releases
+- agent-core v0.4.4 (fixer.js symlink + TOCTOU) -> v0.4.5 (client-side SLSA verification + sync allowlist) synced into all 13 consumers.
+- agent-analyzer v0.8.0 -> v0.8.1 (cargo-deny CI).
+- prepare-delivery v0.1.2, audit-project v1.0.2 (reviewer-contract markers + orchestrator blocked handling).
+
 ## [5.11.0] - 2026-04-26
 
 ### Changed

package/README.md

@@ -19,7 +19,7 @@
 </p>
 
 <p align="center">
-  <b>20 plugins · 49 agents · 41 skills (across all repos) · 30k lines of lib code · 3,507 tests · 5 platforms</b><br>
+  <b>24 plugins · 49 agents · 45 skills (across all repos) · 30k lines of lib code · 3,513 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 - 20 plugins, 49 agents (39 file-based + 10 role-based specialists in audit-project), and 41 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 - 24 plugins, 49 agents (39 file-based + 10 role-based specialists in audit-project), and 45 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.
 
@@ -118,6 +118,8 @@ The investment shifts from model spend to pipeline design. Better prompts, riche
 | [`/next-task`](#next-task) | Task workflow: discovery, implementation, PR, merge |
 | [`/prepare-delivery`](#prepare-delivery) | Pre-ship quality gates: deslop, review, validation, docs sync |
 | [`/gate-and-ship`](#gate-and-ship) | Quality gates then ship (/prepare-delivery + /ship) |
+| [`/axiom`](#axiom) | Durable memory: load, query, list, bootstrap projects, and record approved knowledge |
+| [`/banthis`](#banthis) | Durable negative memory: persist banned agent behaviors |
 | [`/agnix`](#agnix) | Lint agent configurations (399 rules) |
 | [`/ship`](#ship) | PR creation, CI monitoring, merge |
 | [`/deslop`](#deslop) | Clean AI slop patterns |
@@ -133,6 +135,8 @@ The investment shifts from model spend to pipeline design. Better prompts, riche
 | [`/web-ctl`](#web-ctl) | Browser automation for AI agents |
 | [`/release`](#release) | Versioned release with ecosystem detection |
 | [`/skillers`](#skillers) | Workflow pattern learning and automation |
+| [`/skill-curator`](#skill-curator) | Create and improve reliable SKILL.md files |
+| [`/system-prompt-curator`](#system-prompt-curator) | Create and improve autonomous agent system prompts |
 | [`/onboard`](#onboard) | Codebase orientation for newcomers |
 | [`/can-i-help`](#can-i-help) | Match contributor skills to project needs |
 
@@ -142,13 +146,13 @@ Each command works standalone. Together, they compose into end-to-end pipelines.
 
 ## Skills
 
-41 skills included across the plugins:
+45 skills included across the plugins:
 
 | Category | Skills |
 |----------|--------|
 | **Workflow** | `discover-tasks`, `prepare-delivery`, `check-test-coverage`, `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` |
+| **Enhancement** | `enhance-agent-prompts`, `enhance-claude-memory`, `enhance-cross-file`, `enhance-docs`, `enhance-hooks`, `enhance-orchestrator`, `enhance-plugins`, `enhance-prompts`, `enhance-skills`, `skill-curator`, `system-prompt-curator` |
 | **Performance** | `baseline`, `benchmark`, `code-paths`, `investigation-logger`, `perf-analyzer`, `profile`, `theory-gatherer`, `theory-tester` |
 | **Cleanup** | `deslop`, `sync-docs` |
 | **Code Review** | `audit-project` |
@@ -157,6 +161,7 @@ Each command works standalone. Together, they compose into end-to-end pipelines.
 | **Web** | `web-auth`, `web-browse` |
 | **Release** | `release` |
 | **Analysis** | `drift-analysis`, `repo-intel` |
+| **Memory** | `axiom`, `banthis` |
 | **Linting** | `agnix` |
 
 **External skill plugins** (standalone repos, installed separately):
@@ -175,8 +180,8 @@ Skills are the reusable implementation units. Agents invoke skills; commands orc
 |---------|--------------|
 | [The Approach](#the-approach) | Why it's built this way |
 | [Benchmarks](#benchmarks) | Sonnet + agentsys vs raw Opus |
-| [Commands](#commands) | All 20 commands overview |
-| [Skills](#skills) | 41 skills across plugins |
+| [Commands](#commands) | All 24 commands overview |
+| [Skills](#skills) | 45 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 |
@@ -307,6 +312,61 @@ Each piece runs independently - use `/prepare-delivery` alone to review before d
 
 ---
 
+### /axiom
+
+**Purpose:** Durable, queryable memory for agents. Load the smallest useful context, query project or global knowledge, and propose new records without bloating `AGENTS.md`.
+
+**[axiom](https://github.com/agent-sh/axiom)** is a standalone plugin and CLI. It creates a private `axiom-based` knowledge repo after explicit approval, keeps only thin context loaded automatically, and stores durable decisions, memories, preferences, and project notes in queryable files.
+
+**What it does:**
+
+| Command | Use |
+|---------|-----|
+| `axiom before-any --quiet` | Load global thin context at the start of meaningful work |
+| `axiom before-any --project <slug>` | Load project context and create missing project scaffolds |
+| `axiom query "<keyword>" --project <slug>` | Retrieve focused, source-backed project knowledge |
+| `axiom list --topics --project <slug>` | Explore what knowledge exists before querying |
+| `axiom record ...` | Propose a durable record through a temp clone, diff, and human approval |
+
+**Usage:**
+
+```bash
+/axiom before-any --quiet
+/axiom before-any --project flowfabric
+/axiom query "lease based" --project flowfabric
+/axiom record --project flowfabric --kind decision "Lease-based claiming v2" "We switched because it gives stronger safety during restarts."
+```
+
+**External tool:** Requires the [axiom CLI](https://github.com/agent-sh/axiom) from the plugin package.
+
+---
+
+### /banthis
+
+**Purpose:** Durable negative memory for repeated agent mistakes. Turn a user's "stop doing this" correction into a persistent rule in `CLAUDE.md` or `AGENTS.md`.
+
+**[banthis](https://github.com/agent-sh/banthis)** is a tiny standalone CLI plus skill. It maintains a managed banned-behaviors section, supports project or global targets, and includes an `init` meta-rule so agents learn when to invoke it automatically.
+
+**What it does:**
+
+| Command | Use |
+|---------|-----|
+| `banthis add "<title>" "<rule>"` | Add or update a banned behavior |
+| `banthis list` | List current bans |
+| `banthis show` | Print the managed section |
+| `banthis remove "<title>"` | Remove a ban |
+| `banthis init` | Install the meta-rule that teaches agents to call `banthis` |
+
+**Usage:**
+
+```bash
+/banthis "stop ending with vague optional follow-up offers"
+banthis add "No vague endings" "Do not end with vague optional follow-up offers."
+banthis init --file AGENTS.md
+```
+
+---
+
 ### /agnix
 
 **Purpose:** Lint agent configurations before they break your workflow. The first dedicated linter for AI agent configs.
@@ -953,6 +1013,36 @@ No per-turn overhead - it reads transcripts that Claude Code already saves.
 
 ---
 
+### /skill-curator
+
+> Create and improve reliable `SKILL.md` files
+
+```bash
+/skill-curator "create a skill for reviewing background jobs"
+/skill-curator --improve path/to/SKILL.md --category review
+```
+
+The skill curator focuses on trigger quality, clear scope, router patterns, concrete `Skip unless:` gates, token budgets, and agnix-ready structure across Claude Code, Codex, OpenCode, Cursor, Kiro, and similar tools.
+
+**Skill:** skill-curator
+
+---
+
+### /system-prompt-curator
+
+> Create and improve autonomous coding-agent system prompts
+
+```bash
+/system-prompt-curator "GitHub issue resolver"
+/system-prompt-curator --improve path/to/prompt.md --for-orchestrator
+```
+
+The system prompt curator rewrites prompts around task-matched identity, phased workflow, explicit tools, evidence-based completion criteria, and realistic error recovery examples. It separates prompt guidance from harness-level checks that belong in code.
+
+**Skill:** system-prompt-curator
+
+---
+
 ### /onboard
 
 **Purpose:** Get oriented in any codebase in under 3 minutes.
@@ -1237,7 +1327,7 @@ The system is built on research, not guesswork.
 - Instruction following reliability
 
 **Testing:**
-- 3,507 tests passing
+- 3,513 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/index.js

@@ -19,6 +19,22 @@
  *     script validates every zip entry before extracting it and rejects
  *     absolute, UNC, and parent-traversal entries.
  *
+ * Verification chain (in order, each gate must pass to proceed):
+ *   1. TLS - https.get() pins the GitHub CA chain at the OS level.
+ *   2. SHA-256 sidecar - `<asset>.sha256` fetched from the same release and
+ *      verified against the downloaded bytes. Closes basic tampering.
+ *   3. SLSA build provenance (optional / required) - `gh attestation verify`
+ *      checks the Sigstore-signed attestation that agent-analyzer's release
+ *      workflow publishes via `actions/attest-build-provenance`. This closes
+ *      the "stolen release token uploads attacker binary + attacker sha256"
+ *      hole that steps 1 and 2 cannot see.
+ *
+ *      SLSA verification is SOFT by default: if `gh` is not on PATH we log
+ *      a warning and proceed with just SHA-256. Set env var
+ *      `AGENT_ANALYZER_REQUIRE_ATTESTATION=1` to make a missing `gh` a hard
+ *      failure (recommended for CI). A present `gh` that reports a failed
+ *      verification is ALWAYS a hard failure regardless of the env var.
+ *
  * @module lib/binary
  */
 
@@ -572,6 +588,117 @@ function findBinaryInScratch(scratch, binaryBaseName) {
   return null;
 }
 
+// ---------------------------------------------------------------------------
+// SLSA build provenance verification
+// ---------------------------------------------------------------------------
+
+/**
+ * Result of an attempted SLSA attestation verification.
+ * @typedef {Object} SlsaResult
+ * @property {'verified'|'skipped'|'failed'} status
+ * @property {string} [reason]   human-readable detail (for skipped/failed)
+ * @property {string} [stderr]   captured stderr from `gh` (failed only)
+ */
+
+/**
+ * Default runner: spawn `gh attestation verify` and return the captured
+ * exit code, stdout, and stderr. Injectable for tests.
+ * @param {string} filePath
+ * @param {string} repo e.g. `agent-sh/agent-analyzer`
+ * @returns {{ status: number|null, stdout: string, stderr: string }}
+ */
+function defaultGhRunner(filePath, repo) {
+  try {
+    const stdout = cp.execFileSync(
+      'gh',
+      ['attestation', 'verify', filePath, '--repo', repo, '--format', 'json'],
+      {
+        encoding: 'utf8',
+        stdio: ['ignore', 'pipe', 'pipe'],
+        timeout: 60000,
+        windowsHide: true
+      }
+    );
+    return { status: 0, stdout: stdout || '', stderr: '' };
+  } catch (err) {
+    return {
+      status: typeof err.status === 'number' ? err.status : null,
+      stdout: err.stdout ? String(err.stdout) : '',
+      stderr: err.stderr ? String(err.stderr) : (err.message || '')
+    };
+  }
+}
+
+/**
+ * Returns true if the `gh` CLI is on PATH. Uses a short, non-privileged probe.
+ * @param {function} [runner] optional probe; defaults to real `gh --version`
+ * @returns {boolean}
+ */
+function isGhAvailable(runner) {
+  if (typeof runner === 'function') {
+    try { return !!runner(); } catch (e) { return false; }
+  }
+  try {
+    cp.execFileSync('gh', ['--version'], {
+      stdio: 'ignore',
+      timeout: 5000,
+      windowsHide: true
+    });
+    return true;
+  } catch (e) {
+    return false;
+  }
+}
+
+/**
+ * Verify a downloaded asset's SLSA build provenance attestation via the
+ * GitHub CLI. The check is SOFT by default: if `gh` is not installed the
+ * function returns { status: 'skipped' } and the caller logs a warning. Set
+ * `requireAttestation` (or the env var) to make a missing `gh` a failure.
+ *
+ * A present `gh` that reports verification failure ALWAYS returns
+ * { status: 'failed' } regardless of `requireAttestation`; the caller is
+ * expected to abort in that case.
+ *
+ * @param {string} filePath absolute path to the downloaded archive
+ * @param {Object} [options]
+ * @param {string} [options.repo] e.g. `agent-sh/agent-analyzer`
+ * @param {boolean} [options.requireAttestation] defaults to env
+ *   `AGENT_ANALYZER_REQUIRE_ATTESTATION === '1'`
+ * @param {function} [options.ghRunner] injectable runner for tests. Receives
+ *   (filePath, repo), returns { status, stdout, stderr }.
+ * @param {function} [options.ghProbe] injectable gh-on-PATH probe for tests.
+ * @returns {SlsaResult}
+ */
+function verifySlsaAttestation(filePath, options) {
+  const opts = options || {};
+  const repo = opts.repo || GITHUB_REPO;
+  const runner = typeof opts.ghRunner === 'function' ? opts.ghRunner : defaultGhRunner;
+  const require_ = typeof opts.requireAttestation === 'boolean'
+    ? opts.requireAttestation
+    : process.env.AGENT_ANALYZER_REQUIRE_ATTESTATION === '1';
+
+  const ghPresent = isGhAvailable(opts.ghProbe);
+  if (!ghPresent) {
+    const reason = '`gh` CLI not found on PATH';
+    if (require_) {
+      return { status: 'failed', reason: reason + ' (AGENT_ANALYZER_REQUIRE_ATTESTATION=1)' };
+    }
+    return { status: 'skipped', reason: reason };
+  }
+
+  const result = runner(filePath, repo);
+  if (result && result.status === 0) {
+    return { status: 'verified' };
+  }
+  return {
+    status: 'failed',
+    reason: 'gh attestation verify exited with status ' +
+      (result && result.status !== null ? result.status : 'unknown'),
+    stderr: (result && result.stderr) || ''
+  };
+}
+
 // ---------------------------------------------------------------------------
 // Download + install
 // ---------------------------------------------------------------------------
@@ -582,11 +709,19 @@ function findBinaryInScratch(scratch, binaryBaseName) {
  * @param {Object} [options]
  * @param {boolean} [options.skipChecksum=false] LOCAL DEV ONLY. Skips the
  *   `.sha256` sidecar fetch and verification. NEVER set this in production.
+ * @param {boolean} [options.skipAttestation=false] LOCAL DEV ONLY. Skips the
+ *   SLSA attestation check entirely.
+ * @param {boolean} [options.requireAttestation] when true, a missing `gh`
+ *   CLI becomes a hard failure. Defaults to
+ *   `process.env.AGENT_ANALYZER_REQUIRE_ATTESTATION === '1'`.
+ * @param {function} [options.ghRunner] injectable runner for tests.
+ * @param {function} [options.ghProbe] injectable gh-on-PATH probe for tests.
  * @returns {Promise<string>} path to the installed binary
  */
 async function downloadBinary(ver, options) {
   const opts = options || {};
   const skipChecksum = opts.skipChecksum === true;
+  const skipAttestation = opts.skipAttestation === true;
 
   const platformKey = getPlatformKey();
   if (!platformKey) {
@@ -643,6 +778,47 @@ async function downloadBinary(ver, options) {
     verifySha256(buf, expected, filename);
   }
 
+  // --- 2b. Verify SLSA build provenance (optional / required) ------------
+  if (skipAttestation) {
+    process.stderr.write(
+      '[WARN] skipAttestation=true - SLSA verification disabled. ' +
+      'This is LOCAL DEV ONLY and MUST NOT be used in production.\n'
+    );
+  } else {
+    // `gh attestation verify` needs a real file. Persist buf to a tmp path,
+    // verify, then drop it. Extraction continues from the in-memory buf so
+    // we don't need the tmp file beyond the verify call.
+    const attestDir = fs.mkdtempSync(path.join(os.tmpdir(), 'agent-analyzer-slsa-'));
+    const attestFile = path.join(attestDir, filename);
+    try {
+      fs.writeFileSync(attestFile, buf);
+      const result = verifySlsaAttestation(attestFile, {
+        repo: GITHUB_REPO,
+        requireAttestation: opts.requireAttestation,
+        ghRunner: opts.ghRunner,
+        ghProbe: opts.ghProbe
+      });
+      if (result.status === 'verified') {
+        process.stderr.write('[OK] SLSA attestation verified for ' + filename + '\n');
+      } else if (result.status === 'skipped') {
+        process.stderr.write(
+          '[WARN] SLSA attestation check skipped: ' + result.reason + '. ' +
+          'Install the GitHub CLI (`gh`) to enable provenance verification. ' +
+          'Set AGENT_ANALYZER_REQUIRE_ATTESTATION=1 to require it.\n'
+        );
+      } else {
+        // 'failed'
+        throw new Error(
+          'SLSA attestation verification failed for ' + filename + ': ' +
+          result.reason + '. Refusing to execute binary.' +
+          (result.stderr ? '\n--- gh stderr ---\n' + result.stderr : '')
+        );
+      }
+    } finally {
+      rmrf(attestDir);
+    }
+  }
+
   // --- 3. Extract to isolated scratch dir + validate entries -------------
   const binaryBaseName = path.basename(binPath);
   let scratch;
@@ -707,7 +883,13 @@ async function ensureBinary(options) {
     }
   }
 
-  return downloadBinary(targetVer, { skipChecksum: opts.skipChecksum === true });
+  return downloadBinary(targetVer, {
+    skipChecksum: opts.skipChecksum === true,
+    skipAttestation: opts.skipAttestation === true,
+    requireAttestation: opts.requireAttestation,
+    ghRunner: opts.ghRunner,
+    ghProbe: opts.ghProbe
+  });
 }
 
 /**
@@ -730,11 +912,27 @@ function ensureBinarySync(options) {
 
   const targetVer = (options && options.version) || ANALYZER_MIN_VERSION;
   const skipChecksum = !!(options && options.skipChecksum);
+  const skipAttestation = !!(options && options.skipAttestation);
+  // Forward requireAttestation when explicitly set (tri-state: undefined
+  // lets the child fall back to the AGENT_ANALYZER_REQUIRE_ATTESTATION
+  // env var, matching ensureBinary()). Without this forwarding, a sync
+  // caller with requireAttestation:true would silently lose the hard-fail
+  // intent when gh is missing.
+  const requireAttestation = options && typeof options.requireAttestation === 'boolean'
+    ? options.requireAttestation
+    : undefined;
   const selfPath = __filename;
+  const ensureOpts = {
+    version: targetVer,
+    skipChecksum: skipChecksum,
+    skipAttestation: skipAttestation
+  };
+  if (requireAttestation !== undefined) {
+    ensureOpts.requireAttestation = requireAttestation;
+  }
   const helperLines = [
     'var b = require(' + JSON.stringify(selfPath) + ');',
-    'b.ensureBinary({ version: ' + JSON.stringify(targetVer) +
-      ', skipChecksum: ' + JSON.stringify(skipChecksum) + ' })',
+    'b.ensureBinary(' + JSON.stringify(ensureOpts) + ')',
     '  .then(function(p) { process.stdout.write(p); })',
     '  .catch(function(e) { process.stderr.write(e.message); process.exit(1); });'
   ];
@@ -798,6 +996,8 @@ module.exports = {
   assertSafeArchiveEntry,
   assertInsideRoot,
   downloadBinary,
+  verifySlsaAttestation,
+  isGhAvailable,
   // Exported for tests only
   extractTarGzToScratch,
   extractZipToScratch,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agentsys",
-  "version": "5.11.0",
+  "version": "5.13.0",
   "description": "A modular runtime and orchestration system for AI agents - works with Claude Code, OpenCode, and Codex CLI",
   "main": "lib/platform/detect-platform.js",
   "type": "commonjs",

package/scripts/generate-docs.js

@@ -55,11 +55,15 @@ const CATEGORY_MAP = {
   'sync-docs': 'Cleanup',
   'drift-detect': 'Analysis',
   'repo-intel': 'Analysis',
+  'axiom': 'Memory',
+  'banthis': 'Memory',
   'learn': 'AI Collaboration',
   'agnix': 'Linting',
   'consult': 'AI Collaboration',
   'debate': 'AI Collaboration',
   'skillers': 'AI Collaboration',
+  'skill-curator': 'Enhancement',
+  'system-prompt-curator': 'Enhancement',
   'web-ctl': 'Web',
   'ship': 'Release',
   'onboard': 'Onboarding',
@@ -96,12 +100,16 @@ const STATIC_SKILLS = [
   { plugin: 'sync-docs', name: 'sync-docs' },
   { plugin: 'drift-detect', name: 'drift-analysis' },
   { plugin: 'repo-intel', name: 'repo-intel' },
+  { plugin: 'axiom', name: 'axiom' },
+  { plugin: 'banthis', name: 'banthis' },
   { plugin: 'consult', name: 'consult' },
   { plugin: 'debate', name: 'debate' },
   { plugin: 'learn', name: 'learn' },
   { plugin: 'web-ctl', name: 'web-auth' },
   { plugin: 'web-ctl', name: 'web-browse' },
   { plugin: 'ship', name: 'release' },
+  { plugin: 'skill-curator', name: 'skill-curator' },
+  { plugin: 'system-prompt-curator', name: 'system-prompt-curator' },
   { plugin: 'skillers', name: 'skillers-compact' },
   { plugin: 'skillers', name: 'recommend' },
   { plugin: 'onboard', name: 'onboard' },
@@ -126,12 +134,16 @@ const PURPOSE_MAP = {
   'drift-detect': 'Plan drift detection',
   'repo-intel': 'Unified static analysis',
   'sync-docs': 'Documentation sync',
+  'axiom': 'Durable agent-native memory',
+  'banthis': 'Durable negative behavior memory',
   'learn': 'Topic research and learning guides',
   'agnix': 'Agent config linting',
   'consult': 'Cross-tool AI consultation',
   'debate': 'Multi-perspective debate analysis',
   'web-ctl': 'Browser automation for AI agents',
   'skillers': 'Workflow pattern learning',
+  'skill-curator': 'Skill authoring and review',
+  'system-prompt-curator': 'System prompt curation',
   'onboard': 'Codebase onboarding',
   'can-i-help': 'Contributor guidance'
 };
@@ -181,10 +193,11 @@ function generateCommandsTable(commands) {
   // Curated display order (featured commands first, then alphabetical)
   const COMMAND_ORDER = [
     'next-task', 'prepare-delivery', 'gate-and-ship',
-    'agnix', 'ship', 'deslop', 'perf',
+    'axiom', 'banthis', 'agnix', 'ship', 'deslop', 'perf',
     'drift-detect', 'audit-project', 'enhance',
     'repo-intel', 'sync-docs', 'learn', 'consult',
     'debate', 'web-ctl', 'release', 'skillers',
+    'skill-curator', 'system-prompt-curator',
     'onboard', 'can-i-help'
   ];
 
@@ -193,6 +206,8 @@ function generateCommandsTable(commands) {
     'next-task': 'Task workflow: discovery, implementation, PR, merge',
     'prepare-delivery': 'Pre-ship quality gates: deslop, review, validation, docs sync',
     'gate-and-ship': 'Quality gates then ship (/prepare-delivery + /ship)',
+    'axiom': 'Durable memory: load, query, list, bootstrap projects, and record approved knowledge',
+    'banthis': 'Durable negative memory: persist banned agent behaviors',
     'agnix': 'Lint agent configurations (399 rules)',
     'ship': 'PR creation, CI monitoring, merge',
     'deslop': 'Clean AI slop patterns',
@@ -208,6 +223,8 @@ function generateCommandsTable(commands) {
     'web-ctl': 'Browser automation for AI agents',
     'release': 'Versioned release with ecosystem detection',
     'skillers': 'Workflow pattern learning and automation',
+    'skill-curator': 'Create and improve reliable SKILL.md files',
+    'system-prompt-curator': 'Create and improve autonomous agent system prompts',
     'onboard': 'Codebase orientation for newcomers',
     'can-i-help': 'Match contributor skills to project needs'
   };
@@ -267,7 +284,7 @@ function generateSkillsTable(skills) {
   const categoryOrder = [
     'Workflow', 'Message Queues', 'Enhancement', 'Performance', 'Cleanup',
     'Code Review', 'AI Collaboration', 'Onboarding',
-    'Web', 'Release', 'Analysis', 'Linting', 'Other'
+    'Web', 'Release', 'Analysis', 'Memory', 'Linting', 'Other'
   ];
 
   const lines = [
@@ -289,28 +306,36 @@ function generateSkillsTable(skills) {
  * Generate the architecture table for CLAUDE.md / AGENTS.md.
  */
 function generateArchitectureTable(plugins, agents, skills) {
-  const fileBasedAgents = agents.length;
-  const totalAgents = fileBasedAgents + ROLE_BASED_AGENT_COUNT;
-  const totalSkills = skills.length;
+  const effectivePlugins = plugins.length > 0 ? plugins : Object.keys(STATIC_PLUGIN_AGENT_COUNTS);
+  const effectiveSkills = skills.length > 0 ? skills : STATIC_SKILLS;
+  const fileBasedAgents = agents.length > 0 ? agents.length : STATIC_FILE_BASED_AGENT_COUNT;
+  const totalAgents = agents.length > 0
+    ? fileBasedAgents + ROLE_BASED_AGENT_COUNT
+    : STATIC_AGENT_COUNT;
+  const totalSkills = effectiveSkills.length;
 
   // Count agents per plugin (file-based only)
   const agentsByPlugin = {};
-  for (const agent of agents) {
-    agentsByPlugin[agent.plugin] = (agentsByPlugin[agent.plugin] || 0) + 1;
+  if (agents.length > 0) {
+    for (const agent of agents) {
+      agentsByPlugin[agent.plugin] = (agentsByPlugin[agent.plugin] || 0) + 1;
+    }
+  } else {
+    Object.assign(agentsByPlugin, STATIC_PLUGIN_AGENT_COUNTS);
   }
   // audit-project has role-based agents
   agentsByPlugin['audit-project'] = (agentsByPlugin['audit-project'] || 0) + ROLE_BASED_AGENT_COUNT;
 
   // Count skills per plugin
   const skillsByPlugin = {};
-  for (const skill of skills) {
+  for (const skill of effectiveSkills) {
     skillsByPlugin[skill.plugin] = (skillsByPlugin[skill.plugin] || 0) + 1;
   }
 
   const lines = [
     '```',
     'lib/          \u2192 Shared library (vendored to plugins)',
-    `plugins/      \u2192 ${plugins.length} plugins, ${totalAgents} agents (${fileBasedAgents} file-based + ${ROLE_BASED_AGENT_COUNT} role-based), ${totalSkills} skills`,
+    `plugins/      \u2192 ${effectivePlugins.length} plugins, ${totalAgents} agents (${fileBasedAgents} file-based + ${ROLE_BASED_AGENT_COUNT} role-based), ${totalSkills} skills`,
     'adapters/     \u2192 Platform adapters (opencode-plugin/, opencode/, codex/)',
     'checklists/   \u2192 Action checklists (9 files)',
     'bin/cli.js    \u2192 npm CLI installer',
@@ -320,7 +345,7 @@ function generateArchitectureTable(plugins, agents, skills) {
     '|--------|--------|--------|---------|'
   ];
 
-  for (const plugin of plugins) {
+  for (const plugin of effectivePlugins) {
     const agentCount = agentsByPlugin[plugin] || 0;
     const skillCount = skillsByPlugin[plugin] || 0;
     const purpose = PURPOSE_MAP[plugin] || '';
@@ -421,12 +446,16 @@ const STATIC_PLUGIN_AGENT_COUNTS = {
   'enhance': 8,
   'sync-docs': 1,
   'repo-intel': 1,
+  'axiom': 0,
+  'banthis': 0,
   'perf': 6,
   'learn': 1,
   'agnix': 1,
   'consult': 1,
   'debate': 1,
   'web-ctl': 1,
+  'skill-curator': 0,
+  'system-prompt-curator': 0,
   'skillers': 2,
   'onboard': 1,
   'can-i-help': 1,

package/scripts/plugins.txt

@@ -1,5 +1,7 @@
 agnix
+axiom
 audit-project
+banthis
 can-i-help
 consult
 debate
@@ -14,7 +16,9 @@ perf
 prepare-delivery
 repo-intel
 ship
+skill-curator
 skillers
 sync-docs
+system-prompt-curator
 web-ctl
 zig-lsp

package/site/content.json

@@ -1,11 +1,11 @@
 {
   "meta": {
     "title": "agentsys",
-    "description": "A modular runtime and orchestration system for AI agents. 20 plugins, 49 agents, 41 skills - structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.",
+    "description": "A modular runtime and orchestration system for AI agents. 24 plugins, 49 agents, 45 skills - structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.",
     "url": "https://agent-sh.github.io/agentsys",
     "repo": "https://github.com/agent-sh/agentsys",
     "npm": "https://www.npmjs.com/package/agentsys",
-    "version": "5.11.0",
+    "version": "5.13.0",
     "author": "Avi Fenesh",
     "author_url": "https://github.com/avifenesh"
   },
@@ -23,7 +23,7 @@
   },
   "stats": [
     {
-      "value": "20",
+      "value": "24",
       "label": "Plugins",
       "suffix": ""
     },
@@ -33,7 +33,7 @@
       "suffix": ""
     },
     {
-      "value": "41",
+      "value": "45",
       "label": "Skills",
       "suffix": ""
     },
@@ -43,7 +43,7 @@
       "suffix": ""
     },
     {
-      "value": "3,507",
+      "value": "3,513",
       "label": "Tests",
       "suffix": ""
     },
@@ -109,6 +109,20 @@
       "example": "/gate-and-ship",
       "category": "workflow"
     },
+    {
+      "name": "/axiom",
+      "tagline": "Durable memory for agents",
+      "description": "Load thin context, query project or global knowledge, create missing project scaffolds, and propose human-approved records in a private axiom-based knowledge repo.",
+      "example": "/axiom before-any --project flowfabric",
+      "category": "workflow"
+    },
+    {
+      "name": "/banthis",
+      "tagline": "Durable negative memory",
+      "description": "Capture repeated agent mistakes as managed banned-behavior rules in CLAUDE.md or AGENTS.md. Includes a tiny CLI, skill, slash command, and init meta-rule.",
+      "example": "/banthis \"stop ending with vague offers\"",
+      "category": "workflow"
+    },
     {
       "name": "/agnix",
       "tagline": "Lint agent configs before they break",
@@ -214,6 +228,20 @@
       "example": "/skillers compact",
       "category": "learning"
     },
+    {
+      "name": "/skill-curator",
+      "tagline": "Design reliable SKILL.md files",
+      "description": "Create and improve cross-tool skills with strong activation triggers, tight scope, router patterns, Skip unless gates, and agnix-ready structure.",
+      "example": "/skill-curator --improve skills/example/SKILL.md",
+      "category": "maintenance"
+    },
+    {
+      "name": "/system-prompt-curator",
+      "tagline": "Design autonomous agent prompts",
+      "description": "Create and improve system prompts with task-matched identity, phased workflows, explicit tools, evidence-based completion criteria, and error recovery examples.",
+      "example": "/system-prompt-curator \"GitHub issue resolver\"",
+      "category": "maintenance"
+    },
     {
       "name": "/onboard",
       "tagline": "Oriented in any codebase in 3 minutes",
@@ -531,7 +559,7 @@
   ],
   "research": {
     "knowledge_base": "8,000 lines of curated documentation from Anthropic, OpenAI, Google, and Microsoft",
-    "testing": "3,507 tests passing",
+    "testing": "3,513 tests passing",
     "drift_detect_repos": "1,000+ repositories validated",
     "token_reduction": "77% fewer tokens for drift-detect vs multi-agent approaches"
   },

package/site/index.html

@@ -4,12 +4,12 @@
   <meta charset="UTF-8">
   <meta name="viewport" content="width=device-width, initial-scale=1">
   <title>AgentSys - Agent Runtime &amp; Orchestration System</title>
-  <meta name="description" content="A modular runtime and orchestration system for AI agents. 20 plugins, 49 agents, 41 skills - structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.">
+  <meta name="description" content="A modular runtime and orchestration system for AI agents. 24 plugins, 49 agents, 45 skills - structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.">
   <meta name="theme-color" content="#09090b">
 
   <!-- Open Graph -->
   <meta property="og:title" content="AgentSys">
-  <meta property="og:description" content="A modular runtime and orchestration system for AI agents. 20 plugins, 49 agents, 41 skills.">
+  <meta property="og:description" content="A modular runtime and orchestration system for AI agents. 24 plugins, 49 agents, 45 skills.">
   <meta property="og:image" content="https://agent-sh.github.io/agentsys/assets/logo.png">
   <meta property="og:url" content="https://agent-sh.github.io/agentsys/">
   <meta property="og:type" content="website">
@@ -17,7 +17,7 @@
   <!-- Twitter Card -->
   <meta name="twitter:card" content="summary_large_image">
   <meta name="twitter:title" content="AgentSys">
-  <meta name="twitter:description" content="AI workflow automation. 20 plugins, 49 agents, 41 skills.">
+  <meta name="twitter:description" content="AI workflow automation. 24 plugins, 49 agents, 45 skills.">
   <meta name="twitter:image" content="https://agent-sh.github.io/agentsys/assets/logo.png">
 
   <!-- Content Security Policy -->
@@ -107,14 +107,14 @@
       <div class="hero__inner">
         <div class="hero__content">
           <div class="hero__badge anim-fade-in" data-delay="100">
-            20 plugins &middot; 49 agents &middot; 41 skills
+            24 plugins &middot; 49 agents &middot; 45 skills
           </div>
           <h1 class="hero__title anim-fade-up" id="hero-title" data-delay="200">
             A modular <span class="text-gradient">runtime and orchestration system</span><br>
             for AI agents.
           </h1>
           <p class="hero__subtitle anim-fade-up" data-delay="350">
-            Structured pipelines, gated phases, specialized agents. Works with Claude Code, OpenCode, Codex CLI, Cursor, and Kiro. 3,507 tests. Production-grade.
+            Structured pipelines, gated phases, specialized agents. Works with Claude Code, OpenCode, Codex CLI, Cursor, and Kiro. 3,513 tests. Production-grade.
           </p>
           <div class="hero__ctas anim-fade-up" data-delay="500">
             <a href="#install" class="btn btn--primary">Get Started</a>
@@ -156,7 +156,7 @@
     <section class="stats" id="stats" aria-label="Project statistics">
       <div class="stats__inner">
         <div class="stats__item">
-          <span class="stats__number" aria-live="polite" data-target="20">0</span>
+          <span class="stats__number" aria-live="polite" data-target="24">0</span>
           <span class="stats__label">Plugins</span>
         </div>
         <div class="stats__item">
@@ -164,11 +164,11 @@
           <span class="stats__label">Agents</span>
         </div>
         <div class="stats__item">
-          <span class="stats__number" aria-live="polite" data-target="41">0</span>
+          <span class="stats__number" aria-live="polite" data-target="45">0</span>
           <span class="stats__label">Skills</span>
         </div>
         <div class="stats__item">
-          <span class="stats__number" aria-live="polite" data-target="3507">0</span>
+          <span class="stats__number" aria-live="polite" data-target="3513">0</span>
           <span class="stats__label">Tests Passing</span>
         </div>
       </div>
@@ -177,7 +177,7 @@
     <!-- ===== COMMANDS ===== -->
     <section class="commands" id="commands" aria-labelledby="commands-title">
       <div class="commands__inner">
-        <h2 class="commands__title anim-fade-up" id="commands-title">20 Commands. One Toolkit.</h2>
+        <h2 class="commands__title anim-fade-up" id="commands-title">24 Commands. One Toolkit.</h2>
         <p class="commands__subtitle anim-fade-up" data-delay="100">Each works standalone. Together, they automate everything.</p>
 
         <div class="commands__tabs anim-fade-up" data-delay="200">
@@ -202,6 +202,10 @@
             <button class="tabs__tab" role="tab" aria-selected="false" aria-controls="tab-panel-17" id="tab-17" tabindex="-1" data-index="17">/skillers</button>
             <button class="tabs__tab" role="tab" aria-selected="false" aria-controls="tab-panel-18" id="tab-18" tabindex="-1" data-index="18">/onboard</button>
             <button class="tabs__tab" role="tab" aria-selected="false" aria-controls="tab-panel-19" id="tab-19" tabindex="-1" data-index="19">/can-i-help</button>
+            <button class="tabs__tab" role="tab" aria-selected="false" aria-controls="tab-panel-20" id="tab-20" tabindex="-1" data-index="20">/axiom</button>
+            <button class="tabs__tab" role="tab" aria-selected="false" aria-controls="tab-panel-21" id="tab-21" tabindex="-1" data-index="21">/banthis</button>
+            <button class="tabs__tab" role="tab" aria-selected="false" aria-controls="tab-panel-22" id="tab-22" tabindex="-1" data-index="22">/skill-curator</button>
+            <button class="tabs__tab" role="tab" aria-selected="false" aria-controls="tab-panel-23" id="tab-23" tabindex="-1" data-index="23">/system-prompt-curator</button>
           </div>
 
           <!-- Tab panels -->
@@ -584,6 +588,82 @@
 <span class="code-prompt">$</span> /can-i-help --skills=typescript  <span class="code-comment"># Match specific skills</span></code></pre>
             </div>
           </div>
+
+          <div class="tabs__panel" role="tabpanel" id="tab-panel-20" aria-labelledby="tab-20" hidden>
+            <h3 class="tabs__panel-name">/axiom</h3>
+            <p class="tabs__panel-tagline">Durable memory for agents</p>
+            <ul class="tabs__panel-features">
+              <li>Loads thin global and project context</li>
+              <li>Queries durable decisions, memories, and preferences</li>
+              <li>Creates missing project scaffolds automatically</li>
+              <li>Records only through a diff and human approval</li>
+            </ul>
+            <div class="code-block">
+              <button class="code-block__copy" aria-label="Copy code to clipboard" data-code="/axiom before-any --project flowfabric
+/axiom query &quot;lease based&quot; --project flowfabric">
+                <svg width="16" height="16" viewBox="0 0 16 16" fill="currentColor" aria-hidden="true"><path d="M0 6.75C0 5.784.784 5 1.75 5h1.5a.75.75 0 010 1.5h-1.5a.25.25 0 00-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 00.25-.25v-1.5a.75.75 0 011.5 0v1.5A1.75 1.75 0 019.25 16h-7.5A1.75 1.75 0 010 14.25v-7.5z"/><path d="M5 1.75C5 .784 5.784 0 6.75 0h7.5C15.216 0 16 .784 16 1.75v7.5A1.75 1.75 0 0114.25 11h-7.5A1.75 1.75 0 015 9.25v-7.5zm1.75-.25a.25.25 0 00-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 00.25-.25v-7.5a.25.25 0 00-.25-.25h-7.5z"/></svg>
+              </button>
+              <pre><code><span class="code-prompt">$</span> /axiom before-any --project flowfabric
+<span class="code-prompt">$</span> /axiom query "lease based" --project flowfabric</code></pre>
+            </div>
+          </div>
+
+          <div class="tabs__panel" role="tabpanel" id="tab-panel-21" aria-labelledby="tab-21" hidden>
+            <h3 class="tabs__panel-name">/banthis</h3>
+            <p class="tabs__panel-tagline">Durable negative memory</p>
+            <ul class="tabs__panel-features">
+              <li>Persists banned agent behaviors into CLAUDE.md or AGENTS.md</li>
+              <li>Maintains a managed section with add, list, show, remove</li>
+              <li>Includes init meta-rule so agents invoke it automatically</li>
+              <li>Works project-local or global with explicit flags</li>
+            </ul>
+            <div class="code-block">
+              <button class="code-block__copy" aria-label="Copy code to clipboard" data-code="/banthis &quot;stop ending with vague optional offers&quot;
+banthis list">
+                <svg width="16" height="16" viewBox="0 0 16 16" fill="currentColor" aria-hidden="true"><path d="M0 6.75C0 5.784.784 5 1.75 5h1.5a.75.75 0 010 1.5h-1.5a.25.25 0 00-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 00.25-.25v-1.5a.75.75 0 011.5 0v1.5A1.75 1.75 0 019.25 16h-7.5A1.75 1.75 0 010 14.25v-7.5z"/><path d="M5 1.75C5 .784 5.784 0 6.75 0h7.5C15.216 0 16 .784 16 1.75v7.5A1.75 1.75 0 0114.25 11h-7.5A1.75 1.75 0 015 9.25v-7.5zm1.75-.25a.25.25 0 00-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 00.25-.25v-7.5a.25.25 0 00-.25-.25h-7.5z"/></svg>
+              </button>
+              <pre><code><span class="code-prompt">$</span> /banthis "stop ending with vague optional offers"
+<span class="code-prompt">$</span> banthis list</code></pre>
+            </div>
+          </div>
+
+          <div class="tabs__panel" role="tabpanel" id="tab-panel-22" aria-labelledby="tab-22" hidden>
+            <h3 class="tabs__panel-name">/skill-curator</h3>
+            <p class="tabs__panel-tagline">Design reliable SKILL.md files</p>
+            <ul class="tabs__panel-features">
+              <li>Writes strong trigger descriptions and tight scopes</li>
+              <li>Uses router patterns for broad skills</li>
+              <li>Adds concrete Skip unless gates for pattern rules</li>
+              <li>Keeps skills cross-tool and agnix-ready</li>
+            </ul>
+            <div class="code-block">
+              <button class="code-block__copy" aria-label="Copy code to clipboard" data-code="/skill-curator &quot;create a skill for reviewing background jobs&quot;
+/skill-curator --improve path/to/SKILL.md">
+                <svg width="16" height="16" viewBox="0 0 16 16" fill="currentColor" aria-hidden="true"><path d="M0 6.75C0 5.784.784 5 1.75 5h1.5a.75.75 0 010 1.5h-1.5a.25.25 0 00-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 00.25-.25v-1.5a.75.75 0 011.5 0v1.5A1.75 1.75 0 019.25 16h-7.5A1.75 1.75 0 010 14.25v-7.5z"/><path d="M5 1.75C5 .784 5.784 0 6.75 0h7.5C15.216 0 16 .784 16 1.75v7.5A1.75 1.75 0 0114.25 11h-7.5A1.75 1.75 0 015 9.25v-7.5zm1.75-.25a.25.25 0 00-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 00.25-.25v-7.5a.25.25 0 00-.25-.25h-7.5z"/></svg>
+              </button>
+              <pre><code><span class="code-prompt">$</span> /skill-curator "create a skill for reviewing background jobs"
+<span class="code-prompt">$</span> /skill-curator --improve path/to/SKILL.md</code></pre>
+            </div>
+          </div>
+
+          <div class="tabs__panel" role="tabpanel" id="tab-panel-23" aria-labelledby="tab-23" hidden>
+            <h3 class="tabs__panel-name">/system-prompt-curator</h3>
+            <p class="tabs__panel-tagline">Design autonomous agent prompts</p>
+            <ul class="tabs__panel-features">
+              <li>Matches identity to task and tool surface</li>
+              <li>Builds phased workflows with evidence gates</li>
+              <li>Includes realistic error recovery examples</li>
+              <li>Separates prompt guidance from harness checks</li>
+            </ul>
+            <div class="code-block">
+              <button class="code-block__copy" aria-label="Copy code to clipboard" data-code="/system-prompt-curator &quot;GitHub issue resolver&quot;
+/system-prompt-curator --improve prompt.md --for-orchestrator">
+                <svg width="16" height="16" viewBox="0 0 16 16" fill="currentColor" aria-hidden="true"><path d="M0 6.75C0 5.784.784 5 1.75 5h1.5a.75.75 0 010 1.5h-1.5a.25.25 0 00-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 00.25-.25v-1.5a.75.75 0 011.5 0v1.5A1.75 1.75 0 019.25 16h-7.5A1.75 1.75 0 010 14.25v-7.5z"/><path d="M5 1.75C5 .784 5.784 0 6.75 0h7.5C15.216 0 16 .784 16 1.75v7.5A1.75 1.75 0 0114.25 11h-7.5A1.75 1.75 0 015 9.25v-7.5zm1.75-.25a.25.25 0 00-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 00.25-.25v-7.5a.25.25 0 00-.25-.25h-7.5z"/></svg>
+              </button>
+              <pre><code><span class="code-prompt">$</span> /system-prompt-curator "GitHub issue resolver"
+<span class="code-prompt">$</span> /system-prompt-curator --improve prompt.md --for-orchestrator</code></pre>
+            </div>
+          </div>
         </div>
       </div>
     </section>
@@ -662,7 +742,7 @@
     <!-- ===== AGENTS & SKILLS ===== -->
     <section class="agents-skills" id="agents-skills" aria-labelledby="as-title">
       <div class="agents-skills__inner">
-        <h2 class="agents-skills__title anim-fade-up" id="as-title">49 Agents. 41 Skills.</h2>
+        <h2 class="agents-skills__title anim-fade-up" id="as-title">49 Agents. 45 Skills.</h2>
         <p class="agents-skills__subtitle anim-fade-up" data-delay="100">Right model for the task. Opus reasons. Sonnet validates. Haiku executes.</p>
 
         <!-- Agent tier tabs -->
@@ -749,7 +829,7 @@
         </div>
 
         <!-- Skills grid -->
-        <h3 class="agents-skills__skills-title anim-fade-up" data-delay="300">41 Skills across 19 Plugins</h3>
+        <h3 class="agents-skills__skills-title anim-fade-up" data-delay="300">45 Skills across 24 Plugins</h3>
         <div class="skills-grid anim-fade-up" data-delay="350">
           <div class="skill-group">
             <span class="skill-group__label">prepare-delivery</span>
@@ -815,9 +895,13 @@
               <span class="skill-card">repo-intel</span>
               <span class="skill-card">sync-docs</span>
               <span class="skill-card">learn</span>
+              <span class="skill-card">axiom</span>
+              <span class="skill-card">banthis</span>
               <span class="skill-card">consult</span>
               <span class="skill-card">debate</span>
               <span class="skill-card">release</span>
+              <span class="skill-card">skill-curator</span>
+              <span class="skill-card">system-prompt-curator</span>
               <span class="skill-card">onboard</span>
               <span class="skill-card">can-i-help</span>
               <span class="skill-card">audit-project</span>

package/site/ux-spec.md

@@ -94,7 +94,7 @@ Scroll order with rationale for each section. All sections are full-width, alter
 - **Single column on mobile:** Text above, terminal below. Stack with 48px gap.
 
 ### Left Column Content
-1. **Badge** (top, above title): Small pill showing version or "20 plugins . 49 agents . 41 skills"
+1. **Badge** (top, above title): Small pill showing version or "24 plugins . 49 agents . 45 skills"
    - Background: `rgba(99, 102, 241, 0.12)`, border: `1px solid rgba(99, 102, 241, 0.25)`, border-radius: 9999px
    - Font: 13px, font-weight 500, primary accent color
    - Padding: 4px 14px
@@ -104,7 +104,7 @@ Scroll order with rationale for each section. All sections are full-width, alter
    - Color: white
    - "entire dev workflow" portion highlighted with a subtle gradient text (primary-to-secondary accent via `background-clip: text`)
 
-3. **Subtitle:** "20 plugins, 49 agents, 41 skills. From task selection to merged PR. Works with Claude Code, OpenCode, Codex CLI, Cursor, and Kiro."
+3. **Subtitle:** "24 plugins, 49 agents, 45 skills. From task selection to merged PR. Works with Claude Code, OpenCode, Codex CLI, Cursor, and Kiro."
    - Font: 18px on desktop, 16px on mobile, font-weight 400, line-height 1.6
    - Color: `rgba(255, 255, 255, 0.6)`
    - Max-width: 520px
@@ -229,7 +229,7 @@ Done. Task to merged PR in 12 minutes.
 | 1 | 20 | Plugins |
 | 2 | 49 | Agents |
 | 3 | 41 | Skills |
-| 4 | 3,507 | Tests Passing |
+| 4 | 3,513 | Tests Passing |
 
 ### Styling
 - **Number:** 48px, font-weight 700, white, `font-variant-numeric: tabular-nums` (prevents layout shift during count)
@@ -265,7 +265,7 @@ Done. Task to merged PR in 12 minutes.
   - On mobile: naturally scrollable, slight gradient fade on right edge to indicate scrollability (20px gradient from transparent to section background)
 
 ### Tab Order
-`/next-task`, `/ship`, `/deslop`, `/perf`, `/drift-detect`, `/audit-project`, `/enhance`, `/repo-intel`, `/sync-docs`, `/learn`, `/agnix`
+`/next-task`, `/ship`, `/deslop`, `/perf`, `/drift-detect`, `/audit-project`, `/enhance`, `/repo-intel`, `/sync-docs`, `/axiom`, `/learn`, `/agnix`
 
 (Most impactful first: the full workflow command, then the shipper, then individual tools)
 
@@ -650,13 +650,13 @@ This disables:
 ### Head Content
 ```html
 <title>AgentSys - Agent Runtime &amp; Orchestration System</title>
-<meta name="description" content="A modular runtime and orchestration system for AI agents. 20 plugins, 49 agents, 41 skills - structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.">
+<meta name="description" content="A modular runtime and orchestration system for AI agents. 24 plugins, 49 agents, 45 skills - structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.">
 <meta name="viewport" content="width=device-width, initial-scale=1">
 <meta name="theme-color" content="#0a0a0f">
 
 <!-- Open Graph -->
 <meta property="og:title" content="AgentSys">
-<meta property="og:description" content="AI workflow automation. 20 plugins, 49 agents, 41 skills. Task to merged PR.">
+<meta property="og:description" content="AI workflow automation. 24 plugins, 49 agents, 45 skills. Task to merged PR.">
 <meta property="og:image" content="https://agent-sh.github.io/agentsys/assets/og-image.png">
 <meta property="og:url" content="https://agent-sh.github.io/agentsys/">
 <meta property="og:type" content="website">
@@ -664,7 +664,7 @@ This disables:
 <!-- Twitter Card -->
 <meta name="twitter:card" content="summary_large_image">
 <meta name="twitter:title" content="AgentSys">
-<meta name="twitter:description" content="AI workflow automation. 20 plugins, 49 agents, 41 skills.">
+<meta name="twitter:description" content="AI workflow automation. 24 plugins, 49 agents, 45 skills.">
 <meta name="twitter:image" content="https://agent-sh.github.io/agentsys/assets/og-image.png">
 ```