@agentskit/doc-bridge 1.2.1 → 1.2.4

package/docs/index.md

@@ -1,23 +1,58 @@
 ---
 title: Documentation
-description: Choose the shortest Doc Bridge path for humans, agents, or pull-request enforcement.
+description: Practical Doc Bridge paths for humans, agents, PR gates, and MCP — one repository, two audiences.
 ---
 
 # Documentation
 
-Doc Bridge keeps one repository useful to both people and coding agents. Start with the outcome you need:
+Doc Bridge keeps **one repository** useful to people and coding agents. Pick the shortest path:
 
-- **Try it locally:** [Getting started](./getting-started.md)
-- **Route an agent:** [For agents](./for-agents.md)
-- **Block stale context in PRs:** [GitHub Marketplace](./MARKETPLACE.md)
-- **Connect an MCP client:** [MCP](./mcp.md)
+## Start
 
-```mermaid
-flowchart LR
-  D["Repository docs"] --> I["Deterministic index"]
-  I --> H["Human guide"]
-  I --> A["Agent handoff"]
-  I --> G["Pull-request gate"]
+| Goal | Page |
+| --- | --- |
+| Install + 60s proof | [Getting started](./getting-started.md) |
+| Guided install | [Install and run](./guides/install-and-run.md) |
+| Machine-first entry | [For agents](./for-agents.md) · site route `/for-agents` |
+| PR freshness gate | [Gate and CI](./guides/gate-ci.md) · [Marketplace](./MARKETPLACE.md) |
+
+## Build workflows
+
+| Goal | Page |
+| --- | --- |
+| Index + resolve ownership | [Index and query](./guides/index-and-query.md) · [Query](./query.md) |
+| MCP for Cursor / Claude | [MCP for agents](./guides/mcp-agents.md) · [MCP](./mcp.md) |
+| Memory digest → draft docs | [Memory pipeline](./guides/memory-pipeline.md) |
+| Every CLI command | [CLI map](./guides/cli-map.md) · [CLI reference](./spec/cli.md) |
+| Config sketches | [Examples](./examples.md) |
+| Optional chat / RAG | [Chat and RAG](./chat-and-rag.md) · [Ollama demo](./ollama-demo.md) |
+
+## How it works
+
+```text
+Repository docs
+      │
+      ▼
+ak-docs index  ──►  .doc-bridge/index.json + llms.txt
+      │
+      ├─► Human guides (Fumadocs / Docusaurus / md)
+      ├─► Agent handoff (CLI / MCP)
+      └─► PR gate (Marketplace Action)
 ```
 
-Long-form contracts remain available under **Specification** and as raw Markdown, without making the primary path read like a reference manual.
+1. **Index** — map ownership from corpus + config  
+2. **Resolve** — deterministic handoffs (`startHere`, `editRoots`, `checks`)  
+3. **Gate** — fail stale context before agents run  
+4. **Promote** — optional memory → draft docs loop  
+
+## Product vs reference
+
+- **Start / Guides** — get value in minutes  
+- **Product** — [Positioning](./POSITIONING.md), [Playbook pattern](./playbook/doc-bridge-pattern.md), [Recipes](./recipes/index-pipeline.md)  
+- **Reference** — [CLI](./spec/cli.md), [Config](./spec/config-v1.md), [Schemas](./schemas/agent-handoff-v1.md)  
+
+Machine surfaces: [llms.txt](/llms.txt) · [llms-full.txt](/llms-full.txt) · [raw Markdown](/raw/getting-started.md)
+
+## Ecosystem
+
+Part of AgentsKit — next to [AgentsKit](https://www.agentskit.io), [Registry](https://registry.agentskit.io), [Chat](https://chat.agentskit.io), [Playbook](https://playbook.agentskit.io), and [AKOS](https://akos.agentskit.io).

package/docs/mcp.md

@@ -62,3 +62,12 @@ Before editing a package:
 2. Open `startHere`  
 3. Stay inside `editRoots`  
 4. Run `checks` before claiming done  
+
+## Related
+
+- [MCP for agents guide](./guides/mcp-agents.md)  
+- [CLI map](./guides/cli-map.md)  
+- [Memory pipeline](./guides/memory-pipeline.md)  
+- [For agents](./for-agents.md)  
+- [Index and query](./guides/index-and-query.md)  
+- [Skill](./skills/doc-bridge.md) 

package/docs/meta.json

@@ -2,19 +2,25 @@
   "title": "Doc Bridge",
   "pages": [
     "index",
+    "---Start---",
     "getting-started",
+    "guides",
     "for-agents",
     "MARKETPLACE",
-    "POSITIONING",
+    "---Build---",
     "mcp",
     "query",
     "examples",
     "chat-and-rag",
     "ollama-demo",
+    "---Product---",
+    "POSITIONING",
     "playbook",
     "recipes",
+    "---Reference---",
     "schemas",
     "skills",
-    "spec"
+    "spec",
+    "RELEASE"
   ]
 }

package/docs/playbook/doc-bridge-pattern.md

@@ -113,4 +113,4 @@ Teams track handoff % and human-bridge % daily.
 - npm: https://www.npmjs.com/package/@agentskit/doc-bridge
 - repo: https://github.com/AgentsKit-io/doc-bridge
 - skill: [doc-bridge skill](../skills/doc-bridge.md)
-- landing: https://agentskit-io.github.io/doc-bridge/
+- landing: https://doc-bridge.agentskit.io/

package/docs/query.md

@@ -1,34 +1,58 @@
 ---
-title: Deterministic query engine
-description: Ownership and documentation lookup rules for contributors changing Doc Bridge query behavior.
+title: Query
+description: Deterministic ownership and documentation lookup — no model, no re-scan of the repo.
 ---
 
-# Deterministic query engine
+# Query
 
-The query layer reads the generated `.doc-bridge/index.json`; it does not scan
-the repository again and does not call a model. Keep exact ownership resolution,
-ranked document search, and handoff output deterministic.
+The query layer reads `.doc-bridge/index.json` only. It does **not** re-scan the repository and does **not** call a model.
 
-## Ownership
+## Commands
 
-- Source: `src/query/**`
-- Index contract: `src/schemas/doc-bridge-index.ts`
-- Handoff contract: `src/schemas/agent-handoff.ts`
-- CLI consumers: `src/cli/program.ts`
-- MCP consumers: `src/mcp/server.ts`
+```bash
+# Machine-readable handoff (for agents / MCP)
+ak-docs query package auth --agent
+ak-docs query ownership auth --agent
 
-## Before editing
+# Human-readable
+ak-docs query package auth --text
 
-Read [DocBridgeIndex v1](./schemas/doc-bridge-index-v1.md), [AgentHandoff v1](./schemas/agent-handoff-v1.md), and the [CLI reference](./spec/cli.md).
+# Discovery
+ak-docs list packages --text
+ak-docs ask "where do I change billing?"
+```
 
-## Checks
+## What you get
 
-```bash
-pnpm test
-pnpm typecheck
-node bin/ak-docs.js index
-node bin/ak-docs.js query package doc-bridge-query --agent
-```
+An **AgentHandoff** (v1) with stable fields:
+
+| Field | Use |
+| --- | --- |
+| `startHere` | First file the agent should open |
+| `editRoots` | Allowed write paths |
+| `checks` | Verification commands |
+| `humanDoc` | Parallel human documentation |
+
+Schema: [AgentHandoff v1](./schemas/agent-handoff-v1.md) · Index: [DocBridgeIndex v1](./schemas/doc-bridge-index-v1.md)
+
+## When to use which surface
+
+| Need | Surface |
+| --- | --- |
+| Agent about to edit a module | `query … --agent` or MCP `handoff.resolve` |
+| Human browsing ownership | `query … --text` / `list packages` |
+| Free-form question with known docs | `ask "…"` (local first) |
+| Unresolved semantic question | Optional [Chat and RAG](./chat-and-rag.md) backend |
+
+## Guarantees
+
+- Same inputs → same handoff JSON (deterministic)  
+- Breaking field meaning requires a **new schema version**, not silent reinterpretation of v1  
+- Gate/CI can require a fresh index so agents never see stale ownership  
+
+## Related
 
-Preserve stable JSON fields and text output. A breaking contract change requires
-a schema version rather than an implicit reinterpretation of v1.
+- [Guide: Index and query](./guides/index-and-query.md)  
+- [For agents](./for-agents.md)  
+- [CLI reference](./spec/cli.md)  
+- [MCP](./mcp.md)  

package/ecosystem-claims.json

@@ -83,9 +83,9 @@
         },
         {
           "id": "skills",
-          "value": 21,
+          "value": 22,
           "noun": "ready-made skills",
-          "conservativeFloor": 21,
+          "conservativeFloor": 22,
           "evidence": {
             "type": "repository-derivation",
             "repo": "AgentsKit-io/agentskit",
@@ -177,8 +177,8 @@
     {
       "productId": "akos",
       "source": {
-        "type": "endpoint",
-        "url": "https://akos.agentskit.io/api/stats.json"
+        "type": "repository",
+        "repo": "AgentsKit-io/agentskit-os"
       },
       "verification": "declared",
       "claims": []

package/ecosystem-upstream.json

@@ -3,7 +3,7 @@
   "repository": "AgentsKit-io/agentskit",
   "ref": "main",
   "files": {
-    "ecosystem.json": "95adc3c083dd04885e52ff4cb9528b9d73d99d9c69e01408a484e1e862d0e797",
-    "ecosystem-claims.json": "0100f7fbc18a3381bb943aaf8151dd5d107f82c02c470e3c586b7d694c6be42e"
+    "ecosystem.json": "8cc227f6838b9927fcaab5fe019e0f869f8a82bcbb1f2488b8b6b94e41779e9e",
+    "ecosystem-claims.json": "ab8f83b3fa56d525986485dfeb4f12cd89ce1e76b3c90b2b1c6df51dfe8919bd"
   }
 }

package/ecosystem.json

@@ -1,6 +1,6 @@
 {
   "schemaVersion": 2,
-  "$comment": "Canonical registry of AgentsKit ecosystem products. Stable identity lives here; numeric values live in the generated ecosystem-claims ledger. The deprecated properties array temporarily preserves v1 sibling consumers.",
+  "$comment": "Canonical registry of AgentsKit ecosystem products. Stable identity lives in products[] (seven). Numeric values live in the generated ecosystem-claims ledger. The properties array is the full seven-product v1 projection of products[] (same order) for sibling consumers.",
   "parentBrand": {
     "id": "agentskit",
     "name": "AgentsKit"
@@ -16,6 +16,35 @@
       "maturity": "beta",
       "repo": "AgentsKit-io/agentskit",
       "accent": "#2EA043",
+      "showcase": {
+        "claimSource": {
+          "url": "https://www.agentskit.io/api/stats.json",
+          "claims": {
+            "packages": { "path": "counts.packages" },
+            "core-size-kb-gzip": { "path": "coreSizeKbGzip" },
+            "catalog-providers": { "path": "counts.catalogProviders" },
+            "native-adapters": { "path": "counts.nativeAdapters" },
+            "integrations": { "path": "counts.integrations" }
+          }
+        },
+        "stage": "Build",
+        "headline": "One foundation. Every layer stays yours.",
+        "detail": "Compose runtime, adapters, tools, memory, RAG, and UI without glue code or lock-in.",
+        "proof": "{{packages}} packages · <{{core-size-kb-gzip}} KB core",
+        "sales": {
+          "kind": "integration-stack",
+          "headline": "Swap the stack. Keep the agent.",
+          "metrics": [
+            { "value": "{{catalog-providers}}", "label": "providers" },
+            { "value": "{{native-adapters}}", "label": "adapters" },
+            { "value": "{{integrations}}", "label": "integrations" }
+          ],
+          "capabilities": ["Tools", "RAG", "Memory", "MCP"],
+          "steps": ["Choose any adapter", "Add tools and memory", "Ship without rewrites"]
+        },
+        "cta": "Build with AgentsKit",
+        "ctaSurface": "docs"
+      },
       "surfaces": {
         "home": "https://www.agentskit.io",
         "docs": "https://www.agentskit.io/docs",
@@ -27,7 +56,14 @@
       "navigation": {
         "showInBar": true,
         "order": 0,
-        "next": ["registry", "agentskit-chat", "playbook", "doc-bridge", "code-review", "akos"]
+        "next": [
+          "registry",
+          "agentskit-chat",
+          "playbook",
+          "doc-bridge",
+          "code-review",
+          "akos"
+        ]
       }
     },
     {
@@ -40,6 +76,30 @@
       "maturity": "beta",
       "repo": "AgentsKit-io/agentskit-registry",
       "accent": "#58A6FF",
+      "showcase": {
+        "claimSource": {
+          "url": "https://registry.agentskit.io/r/index.json",
+          "claims": {
+            "agents": { "path": "agents", "aggregate": "length" },
+            "remaining-categories": { "path": "agents", "aggregate": "distinct", "field": "category", "subtract": 5 }
+          }
+        },
+        "stage": "Discover",
+        "headline": "Shadcn-like agents. Installed as source.",
+        "detail": "Find a working agent, copy its source into your project, and change every line.",
+        "proof": "Ready-made · source-owned",
+        "sales": {
+          "kind": "registry-install",
+          "headline": "Ready-to-use agents from the live Registry.",
+          "metric": "{{agents}}",
+          "metricLabel": "shadcn-like agents",
+          "capabilities": ["Research", "Support", "Coding", "Data", "Marketing", "{{remaining-categories}} additional categories"],
+          "steps": ["Find the right agent", "Run npx agentskit add", "Own every line"],
+          "command": "npx agentskit add research"
+        },
+        "cta": "Explore the Registry",
+        "ctaSurface": "home"
+      },
       "surfaces": {
         "home": "https://registry.agentskit.io",
         "docs": "https://registry.agentskit.io/docs",
@@ -51,7 +111,14 @@
       "navigation": {
         "showInBar": true,
         "order": 1,
-        "next": ["agentskit", "agentskit-chat", "playbook", "doc-bridge", "code-review", "akos"]
+        "next": [
+          "agentskit",
+          "agentskit-chat",
+          "playbook",
+          "doc-bridge",
+          "code-review",
+          "akos"
+        ]
       }
     },
     {
@@ -64,6 +131,28 @@
       "maturity": "alpha",
       "repo": "AgentsKit-io/agentskit-chat",
       "accent": "#F59E0B",
+      "showcase": {
+        "stage": "Deliver",
+        "headline": "One agent. Every conversation surface.",
+        "detail": "Define the experience once and deliver it across web, desktop, CLI, and mobile.",
+        "proof": "Web · Desktop · CLI · Mobile",
+        "sales": {
+          "kind": "human-agent",
+          "headline": "Human ↔ agent. Without losing control.",
+          "metric": "4",
+          "metricLabel": "surfaces · one conversation",
+          "logos": [
+            { "id": "web", "label": "Web" },
+            { "id": "desktop", "label": "Desktop" },
+            { "id": "cli", "label": "CLI" },
+            { "id": "mobile", "label": "Mobile" }
+          ],
+          "capabilities": ["Streaming", "Approvals", "Tool results", "Shared state"],
+          "steps": ["Agent requests a protected action", "Human approves", "The conversation continues everywhere"]
+        },
+        "cta": "Explore AgentsKit Chat",
+        "ctaSurface": "home"
+      },
       "surfaces": {
         "home": "https://chat.agentskit.io",
         "docs": "https://chat.agentskit.io/docs",
@@ -74,7 +163,14 @@
       "navigation": {
         "showInBar": true,
         "order": 2,
-        "next": ["agentskit", "registry", "playbook", "doc-bridge", "code-review", "akos"]
+        "next": [
+          "agentskit",
+          "registry",
+          "playbook",
+          "doc-bridge",
+          "code-review",
+          "akos"
+        ]
       }
     },
     {
@@ -83,10 +179,32 @@
       "shortName": "Playbook",
       "kind": "methodology",
       "role": "discipline",
-      "promise": "Make agents ship code a human would actually merge.",
+      "promise": "Train coding agents to ship code your team would merge.",
       "maturity": "stable",
       "repo": "AgentsKit-io/agents-playbook",
       "accent": "#8B5CF6",
+      "showcase": {
+        "stage": "Standardize",
+        "headline": "The open engineering harness for coding agents.",
+        "detail": "Open rules, prompts, memory, evals, and executable gates turn every correction into behavior coding agents can repeat.",
+        "proof": "Train the behavior · not the model",
+        "sales": {
+          "kind": "standards-flow",
+          "headline": "Your standards. Every agent. Every repository.",
+          "metric": "1×",
+          "metricLabel": "define · reuse everywhere",
+          "logos": [
+            { "id": "architecture", "label": "Architecture" },
+            { "id": "tests", "label": "Tests" },
+            { "id": "security", "label": "Security" },
+            { "id": "style", "label": "Style" }
+          ],
+          "capabilities": ["Repository rules", "Review criteria", "Agent guidance", "Team conventions"],
+          "steps": ["Define the standard once", "Agents receive executable guidance", "Review consistent output"]
+        },
+        "cta": "Explore the harness",
+        "ctaSurface": "home"
+      },
       "surfaces": {
         "home": "https://playbook.agentskit.io",
         "docs": "https://playbook.agentskit.io/docs",
@@ -98,7 +216,14 @@
       "navigation": {
         "showInBar": true,
         "order": 3,
-        "next": ["agentskit", "registry", "agentskit-chat", "doc-bridge", "code-review", "akos"]
+        "next": [
+          "agentskit",
+          "registry",
+          "agentskit-chat",
+          "doc-bridge",
+          "code-review",
+          "akos"
+        ]
       }
     },
     {
@@ -111,17 +236,46 @@
       "maturity": "stable",
       "repo": "AgentsKit-io/doc-bridge",
       "accent": "#06B6D4",
+      "showcase": {
+        "stage": "Understand",
+        "headline": "Documentation that hands work off precisely.",
+        "detail": "Connect repository knowledge to agents with structured, executable context.",
+        "proof": "Code → precise handoff",
+        "sales": {
+          "kind": "knowledge-bridge",
+          "headline": "Knowledge flows both ways.",
+          "metric": "↔",
+          "metricLabel": "humans and agents",
+          "logos": [
+            { "id": "human", "label": "Humans" },
+            { "id": "markdown", "label": "Docs" },
+            { "id": "code", "label": "Code" },
+            { "id": "agent", "label": "Agents" }
+          ],
+          "capabilities": ["ADRs", "Handoffs", "Agent findings", "Human-readable docs"],
+          "steps": ["Humans document decisions", "Agents receive precise context", "Agent findings return to humans"]
+        },
+        "cta": "Generate your first handoff",
+        "ctaSurface": "home"
+      },
       "surfaces": {
-        "home": "https://agentskit-io.github.io/doc-bridge/",
-        "docs": "https://agentskit-io.github.io/doc-bridge/",
-        "llms": "https://agentskit-io.github.io/doc-bridge/llms.txt",
+        "home": "https://doc-bridge.agentskit.io/",
+        "docs": "https://doc-bridge.agentskit.io/",
+        "llms": "https://doc-bridge.agentskit.io/llms.txt",
         "documentation": "fumadocs",
         "chat": "none"
       },
       "navigation": {
         "showInBar": true,
         "order": 4,
-        "next": ["agentskit", "registry", "agentskit-chat", "playbook", "code-review", "akos"]
+        "next": [
+          "agentskit",
+          "registry",
+          "agentskit-chat",
+          "playbook",
+          "code-review",
+          "akos"
+        ]
       }
     },
     {
@@ -142,9 +296,16 @@
         "chat": "none"
       },
       "navigation": {
-        "showInBar": true,
+        "showInBar": false,
         "order": 5,
-        "next": ["agentskit", "registry", "agentskit-chat", "playbook", "doc-bridge", "akos"]
+        "next": [
+          "agentskit",
+          "registry",
+          "agentskit-chat",
+          "playbook",
+          "doc-bridge",
+          "akos"
+        ]
       }
     },
     {
@@ -157,11 +318,32 @@
       "maturity": "stable",
       "repo": "AgentsKit-io/agentskit-os",
       "accent": "#34D399",
+      "showcase": {
+        "stage": "Operate",
+        "headline": "Run and govern agents in production.",
+        "detail": "Add orchestration, governance, and operational control when your system needs it.",
+        "proof": "Orchestration · governance · control",
+        "sales": {
+          "kind": "enterprise-control",
+          "headline": "The enterprise control plane for agents.",
+          "metric": "Enterprise",
+          "metricLabel": "operations and governance",
+          "logos": [
+            { "id": "logs", "label": "Logs" },
+            { "id": "approvals", "label": "Approvals" },
+            { "id": "rbac", "label": "RBAC" },
+            { "id": "policies", "label": "Policies" }
+          ],
+          "capabilities": ["Logs", "Approvals", "RBAC", "Policies", "Tracing", "Costs"],
+          "steps": ["Agent requests a protected action", "RBAC and policy gates evaluate", "Human approves; logs and trace persist"]
+        },
+        "cta": "Explore AgentsKit OS",
+        "ctaSurface": "home"
+      },
       "surfaces": {
         "home": "https://akos.agentskit.io",
         "docs": "https://akos.agentskit.io/docs",
         "llms": "https://akos.agentskit.io/llms.txt",
-        "stats": "https://akos.agentskit.io/api/stats.json",
         "documentation": "fumadocs",
         "chat": "none"
       },
@@ -187,17 +369,29 @@
       "stats": "https://www.agentskit.io/api/stats.json"
     },
     {
-      "id": "akos",
-      "name": "AgentsKit OS",
-      "barLabel": "AKOS",
-      "domain": "akos.agentskit.io",
-      "url": "https://akos.agentskit.io",
-      "repo": "AgentsKit-io/agentskit-os",
-      "tagline": "Run and govern agents in production.",
-      "kind": "product",
-      "accent": "#34D399",
-      "llms": "https://akos.agentskit.io/llms.txt",
-      "stats": "https://akos.agentskit.io/api/stats.json"
+      "id": "registry",
+      "name": "AgentsKit Registry",
+      "barLabel": "Registry",
+      "domain": "registry.agentskit.io",
+      "url": "https://registry.agentskit.io",
+      "repo": "AgentsKit-io/agentskit-registry",
+      "tagline": "Copy ready-made agents and own the source.",
+      "kind": "registry",
+      "accent": "#58A6FF",
+      "llms": "https://registry.agentskit.io/llms.txt",
+      "stats": "https://registry.agentskit.io/r/index.json"
+    },
+    {
+      "id": "agentskit-chat",
+      "name": "AgentsKit Chat",
+      "barLabel": "Chat",
+      "domain": "chat.agentskit.io",
+      "url": "https://chat.agentskit.io",
+      "repo": "AgentsKit-io/agentskit-chat",
+      "tagline": "Define one agent experience and deliver it across interfaces.",
+      "kind": "framework",
+      "accent": "#F59E0B",
+      "llms": "https://chat.agentskit.io/llms.txt"
     },
     {
       "id": "playbook",
@@ -206,24 +400,47 @@
       "domain": "playbook.agentskit.io",
       "url": "https://playbook.agentskit.io",
       "repo": "AgentsKit-io/agents-playbook",
-      "tagline": "Make agents ship code a human would actually merge.",
+      "tagline": "Train coding agents to ship code your team would merge.",
       "kind": "methodology",
       "accent": "#8B5CF6",
       "llms": "https://playbook.agentskit.io/llms.txt",
       "stats": "https://playbook.agentskit.io/api/stats.json"
     },
     {
-      "id": "registry",
-      "name": "AgentsKit Registry",
-      "barLabel": "Registry",
-      "domain": "registry.agentskit.io",
-      "url": "https://registry.agentskit.io",
-      "repo": "AgentsKit-io/agentskit-registry",
-      "tagline": "Copy ready-made agents and own the source.",
-      "kind": "registry",
-      "accent": "#58A6FF",
-      "llms": "https://registry.agentskit.io/llms.txt",
-      "stats": "https://registry.agentskit.io/r/index.json"
+      "id": "doc-bridge",
+      "name": "Doc Bridge",
+      "barLabel": "Doc Bridge",
+      "domain": "doc-bridge.agentskit.io",
+      "url": "https://doc-bridge.agentskit.io/",
+      "repo": "AgentsKit-io/doc-bridge",
+      "tagline": "Turn repository documentation into executable agent handoffs.",
+      "kind": "developer-tool",
+      "accent": "#06B6D4",
+      "llms": "https://doc-bridge.agentskit.io/llms.txt"
+    },
+    {
+      "id": "code-review",
+      "name": "AgentsKit Code Review",
+      "barLabel": "Code Review",
+      "domain": "github.com",
+      "url": "https://github.com/AgentsKit-io/code-review-cli",
+      "repo": "AgentsKit-io/code-review-cli",
+      "tagline": "Run deep, low-noise review with the model already in use.",
+      "kind": "developer-tool",
+      "accent": "#F97316",
+      "llms": "https://raw.githubusercontent.com/AgentsKit-io/code-review-cli/main/llms.txt"
+    },
+    {
+      "id": "akos",
+      "name": "AgentsKit OS",
+      "barLabel": "AKOS",
+      "domain": "akos.agentskit.io",
+      "url": "https://akos.agentskit.io",
+      "repo": "AgentsKit-io/agentskit-os",
+      "tagline": "Run and govern agents in production.",
+      "kind": "product",
+      "accent": "#34D399",
+      "llms": "https://akos.agentskit.io/llms.txt"
     }
   ],
   "builder": {