omnius 1.0.591 → 1.0.592

package/.aiwg/addons/omnius-docs/README.md

@@ -2,6 +2,20 @@
 
 This project-local addon exposes Omnius documentation as small discoverable skills. It is for agents that need to answer questions about Omnius without loading the entire docs tree.
 
-Start with `omnius-docs`, then load a more specific skill when the task is about realtime, sponsors, Telegram, operations, or browser interaction validation.
+Start with `omnius-docs`, then load a more specific skill when the task is
+about inference providers, tools/web search, service integration, version
+compatibility, realtime, sponsors, Telegram, operations, or browser
+interaction validation.
+
+Documentation-routing skills:
+
+- `omnius-inference-docs`
+- `omnius-tools-docs`
+- `omnius-integration-docs`
+- `omnius-version-compatibility-docs`
+- `omnius-realtime-docs`
+- `omnius-sponsor-docs`
+- `omnius-telegram-docs`
+- `omnius-ops-docs`
 
 For cross-cutting delivery work, the add-on also exposes `evidence-directed-delivery`, `workspace-reality-audit`, `test-surface-audit`, `runtime-provenance-audit`, `hardware-evidence-audit`, `secrets-and-config-audit`, and `agent-failure-recovery`. These skills are feedback-oriented: they help an agent discover evidence gaps and recover from actual failures without treating planning telemetry as a completion veto.

package/.aiwg/addons/omnius-docs/manifest.json

@@ -1,72 +1,32 @@
 {
-  "name": "omnius-docs",
-  "version": "1.0.0",
+  "id": "omnius-docs",
+  "type": "addon",
+  "name": "Omnius Documentation",
+  "version": "1.1.0",
   "description": "Project-local Omnius documentation bundle for discover-then-expand agent workflows.",
-  "skills": [
-    {
-      "name": "omnius-docs",
-      "description": "Find the right Omnius documentation page for setup, TUI, endpoints, realtime, sponsor, COHERE, Telegram, media, operations, architecture, and REST references.",
-      "triggers": ["Omnius docs", "Omnius documentation", "how to use Omnius", "Omnius guide", "Omnius manual"]
-    },
-    {
-      "name": "omnius-realtime-docs",
-      "description": "Use Omnius realtime spoken conversation mode through /realtime or REST realtime=true.",
-      "triggers": ["Omnius realtime", "/realtime", "realtime REST", "ASR TTS chat", "voice conversation mode"]
-    },
-    {
-      "name": "omnius-sponsor-docs",
-      "description": "Configure and consume sponsor inference, COHERE distributed inference, endpoint passthrough, media sponsorship, and peer telemetry.",
-      "triggers": ["Omnius sponsor", "/sponsor", "/cohere", "sponsor endpoint", "COHERE mesh", "media sponsorship"]
-    },
-    {
-      "name": "omnius-telegram-docs",
-      "description": "Configure Telegram bridge behavior, scoped persona state, reply preferences, media intake, and raw failure feedback.",
-      "triggers": ["Omnius Telegram", "/telegram", "Telegram persona", "Telegram reply preferences", "Telegram failure feedback"]
-    },
-    {
-      "name": "omnius-ops-docs",
-      "description": "Operate Omnius safely: .omnius hygiene, Ollama cleanup, broker, remote access, API keys, elevation prompts, and runtime failure capture.",
-      "triggers": ["Omnius operations", "Ollama cleanup", "/broker", "remote access", "API keys", ".omnius"]
-    },
-    {
-      "name": "browser-interaction-validation",
-      "description": "Validate a user-facing browser flow with canonical Playwright interactions, postconditions, diagnostics, and honest blocked/not-applicable evidence.",
-      "triggers": ["browser interaction validation", "UI interaction test", "Playwright validation", "browser flow", "dashboard validation", "form interaction"]
-    },
-    {
-      "name": "evidence-directed-delivery",
-      "description": "Use generic claim/evidence, anti-reward-hacking, bounded recovery, and feedback-only completion rules for non-trivial work.",
-      "triggers": ["evidence based completion", "delivery audit", "completion evidence", "verify claims", "remaining gaps"]
-    },
-    {
-      "name": "workspace-reality-audit",
-      "description": "Audit repository truth, generated artifacts, documentation drift, and delivery ownership before claiming progress.",
-      "triggers": ["workspace audit", "repository reality", "stale artifact", "untracked source", "documentation drift"]
-    },
-    {
-      "name": "test-surface-audit",
-      "description": "Map canonical tests to claims and expose placeholder, mock-only, copied-logic, and un-wired verification gaps.",
-      "triggers": ["test audit", "test coverage", "test surface", "canonical test", "mock only"]
-    },
-    {
-      "name": "runtime-provenance-audit",
-      "description": "Prove active process, artifact, port, configuration, and restart/update ownership for runtime claims.",
-      "triggers": ["runtime provenance", "deployed version", "process audit", "service ownership", "running artifact"]
-    },
-    {
-      "name": "hardware-evidence-audit",
-      "description": "Classify static, simulation, integration, and physical-device evidence without overstating hardware proof.",
-      "triggers": ["hardware validation", "device evidence", "simulation versus hardware", "firmware test", "physical device"]
-    },
-    {
-      "name": "secrets-and-config-audit",
-      "description": "Audit credential and configuration provenance without exposing secret values.",
-      "triggers": ["secret audit", "credential provenance", "endpoint credential", "configuration security"]
-    },
-    {
-      "name": "agent-failure-recovery",
-      "description": "Convert failed calls into bounded, evidence-backed recovery actions and stop blind retries.",
-      "triggers": ["recover tool failure", "stale hash", "unknown tool", "retry loop", "agent recovery"]
+  "manifestVersion": "1",
+  "platforms": {
+    "claude": "full",
+    "codex": "full",
+    "factory": "full",
+    "cursor": "full",
+    "generic": "full"
+  },
+  "keywords": [
+    "omnius",
+    "documentation",
+    "discovery",
+    "inference",
+    "tools",
+    "operations",
+    "project-local"
+  ],
+  "deployment": {
+    "pathTemplate": ".{platform}/skills/{id}.md"
+  },
+  "addonConfig": {
+    "entry": {
+      "skills": "skills/"
     }
-  ]
+  }
 }

package/.aiwg/addons/omnius-docs/skills/agent-failure-recovery/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: agent-failure-recovery
+description: Convert failed calls into bounded evidence-backed recovery actions and stop blind retries.
 compaction_strategy: structured
 ---
 
@@ -20,4 +22,3 @@ timeout, or contradictory evidence.
 5. Resolve the failure only with evidence that addresses the same tool family,
    claim, and current state. Keep it as feedback; do not use it as a terminal
    completion gate.
-

package/.aiwg/addons/omnius-docs/skills/browser-interaction-validation/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: browser-interaction-validation
+description: Validate a user-facing browser flow with canonical Playwright interactions, explicit postconditions, and honest blocked or inconclusive evidence.
 compaction_strategy: structured
 ---
 
@@ -78,4 +80,3 @@ browser flow: fleet control
   user gesture; no physical device is attached to this browser session.
 - static evidence: dashboard harness passed (does not prove the three flows).
 ```
-

package/.aiwg/addons/omnius-docs/skills/evidence-directed-delivery/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: evidence-directed-delivery
+description: Use claim-and-evidence, anti-reward-hacking, bounded recovery, and feedback-only completion rules for non-trivial work.
 compaction_strategy: structured
 ---
 
@@ -60,4 +62,3 @@ recovery: stale edit target
 - status: resolved
 - evidence: re-read current target, rebuilt the edit, then ran a fresh verifier
 ```
-

package/.aiwg/addons/omnius-docs/skills/hardware-evidence-audit/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: hardware-evidence-audit
+description: Classify static, simulation, integration, and physical-device evidence without overstating hardware proof.
 compaction_strategy: structured
 ---
 
@@ -22,4 +24,3 @@ Never promote an item to a higher level without a fresh observation at that
 level. Before live local inference or hardware work, follow the project’s
 hardware preflight rules. If the required device, endpoint, permission, or
 accelerator cannot be proven, report `blocked` or `not_applicable`.
-

package/.aiwg/addons/omnius-docs/skills/omnius-docs/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: omnius-docs
+description: Find the smallest Omnius documentation artifact for setup, workflows, providers, tools, APIs, operations, or architecture.
 compaction_strategy: structured
 ---
 
@@ -19,18 +21,24 @@ Find and use Omnius documentation without loading the whole docs tree.
 ## Workflow
 
 1. Start with `README.md` only for product overview and top-level docs map.
-2. For setup questions, open `docs/getting-started/`.
-3. For user workflows, open the relevant file under `docs/guides/`.
-4. For commands, config, and API inventory, open `docs/reference/`.
-5. For runtime and deployment questions, open `docs/operations/`.
-6. For implementation shape, open `docs/architecture/overview.md`.
-7. Read only the smallest relevant page before answering.
+2. Search `docs/DISCOVERY.json` by stable ID, alias, or intent.
+3. For setup questions, open `docs/getting-started/`.
+4. For user workflows, open the relevant file under `docs/guides/`.
+5. For commands, config, and API inventory, open `docs/reference/`.
+6. For runtime and deployment questions, open `docs/operations/`.
+7. For implementation shape, open `docs/architecture/overview.md`.
+8. Read only the smallest relevant page before answering.
 
 ## Map
 
 - Install: `docs/getting-started/install.md`
 - First run: `docs/getting-started/first-run.md`
 - Model providers: `docs/getting-started/model-providers.md`
+- Discovery cascade: `docs/DISCOVERY.md`
+- Bring your own inference: `docs/guides/bring-your-own-inference.md`
+- Tools and web search: `docs/guides/tools-and-web-search.md`
+- Agent integration: `docs/guides/agent-integration.md`
+- Service compatibility: `docs/operations/version-compatibility.md`
 - Realtime: `docs/guides/realtime.md`
 - Sponsor and COHERE: `docs/guides/sponsor-and-cohere.md`
 - Telegram: `docs/guides/telegram.md`
@@ -45,4 +53,6 @@ Find and use Omnius documentation without loading the whole docs tree.
 
 ## Canonical Sources
 
-When documentation and code disagree, verify against the code or live daemon. REST machine contract lives in `packages/cli/src/api/openapi.ts`.
+When documentation and code disagree, verify against the code or live daemon.
+REST machine contract lives in `packages/cli/src/api/openapi.ts`; offline
+discovery lives in `docs/DISCOVERY.json`.

package/.aiwg/addons/omnius-docs/skills/omnius-inference-docs/SKILL.md

@@ -0,0 +1,27 @@
+---
+name: omnius-inference-docs
+description: Discover Omnius provider protocols, endpoints, credentials, Claude Messages translation, Gemini compatibility, and fail-closed custom endpoint behavior.
+compaction_strategy: structured
+---
+
+# omnius-inference-docs
+
+Use this skill for bring-your-own inference, provider accounts, endpoint
+protocols, provider keys, Claude, Gemini, and custom endpoints.
+
+## Workflow
+
+1. Read `docs/guides/bring-your-own-inference.md`.
+2. Expand the provider's stable entry in `docs/DISCOVERY.json`.
+3. Read `docs/reference/configuration.md` for key precedence.
+4. For REST adapters, also read
+   `docs/operations/version-compatibility.md`.
+5. Verify live provider and version metadata before execution.
+
+## Invariants
+
+- Anthropic/Claude uses `anthropic-messages`, not a guessed OpenAI path.
+- Gemini uses the official `/v1beta/openai` compatibility prefix.
+- Unknown custom protocols fail closed without a network request.
+- REST bearer keys and provider credentials are separate.
+- Local inference requires the repository hardware preflight.

package/.aiwg/addons/omnius-docs/skills/omnius-integration-docs/SKILL.md

@@ -0,0 +1,21 @@
+---
+name: omnius-integration-docs
+description: Integrate an agent or service with Omnius through discovery, OpenAPI, provider descriptors, live tool metadata, scoped auth, and a version precondition.
+compaction_strategy: structured
+---
+
+# omnius-integration-docs
+
+Use when another agent, account adapter, service, or SDK needs to call Omnius.
+
+## Workflow
+
+1. Read `docs/guides/agent-integration.md`.
+2. Inspect `/version` and select the minimum compatible package version.
+3. Discover a stable capability ID and expand its interfaces.
+4. Validate provider protocol or live tool invocation metadata.
+5. Send `X-Omnius-Min-Version` on every execution request.
+6. Keep provider credentials separate from scoped daemon bearer keys.
+
+Discovery describes the available contract; live OpenAPI and policy state
+remain authoritative for the running service.

package/.aiwg/addons/omnius-docs/skills/omnius-ops-docs/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: omnius-ops-docs
+description: Operate Omnius safely through runtime hygiene, remote access, scoped keys, model cleanup, and failure capture guidance.
 compaction_strategy: structured
 ---
 

package/.aiwg/addons/omnius-docs/skills/omnius-realtime-docs/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: omnius-realtime-docs
+description: Use Omnius realtime spoken conversation through the TUI or REST with bounded ASR, TTS, history, and persona context.
 compaction_strategy: structured
 ---
 

package/.aiwg/addons/omnius-docs/skills/omnius-sponsor-docs/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: omnius-sponsor-docs
+description: Configure sponsor and COHERE inference, endpoint passthrough, media sponsorship, and peer telemetry.
 compaction_strategy: structured
 ---
 

package/.aiwg/addons/omnius-docs/skills/omnius-telegram-docs/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: omnius-telegram-docs
+description: Configure the Telegram bridge, scoped persona state, reply preferences, media intake, and failure feedback.
 compaction_strategy: structured
 ---
 

package/.aiwg/addons/omnius-docs/skills/omnius-tools-docs/SKILL.md

@@ -0,0 +1,23 @@
+---
+name: omnius-tools-docs
+description: Discover Omnius tools and select the correct direct, agent-loop, external HTTP, or MCP invocation interface.
+compaction_strategy: structured
+---
+
+# omnius-tools-docs
+
+Use for built-in tools, web search, direct tool calls, external tools, MCP,
+tool security, and tool profiles.
+
+## Workflow
+
+1. Read `docs/guides/tools-and-web-search.md`.
+2. Inspect the stable catalog entry and live `GET /v1/tools/{name}` metadata.
+3. Use a direct call only when `direct_callable` is true.
+4. Route agent-bound tools through `/v1/run`, `/v1/chat`, or
+   `/v1/chat/completions`.
+5. Preserve auth scope, profile, origin, network, and risk gates.
+
+`web_search` is agent-bound. Its schema can be discovered, but it should be
+offered to an Omnius agent loop instead of assumed callable at the direct tool
+route.

package/.aiwg/addons/omnius-docs/skills/omnius-version-compatibility-docs/SKILL.md

@@ -0,0 +1,23 @@
+---
+name: omnius-version-compatibility-docs
+description: Prevent a stale Omnius daemon from executing jobs by inspecting version metadata and sending the X-Omnius-Min-Version precondition.
+compaction_strategy: structured
+---
+
+# omnius-version-compatibility-docs
+
+Use for client/daemon compatibility, stale services, package version checks,
+and HTTP `400` / `412` version failures.
+
+## Workflow
+
+1. Read `docs/operations/version-compatibility.md`.
+2. Inspect `GET /version`.
+3. Compare `package_version` with the integration's required SemVer.
+4. Send `X-Omnius-Min-Version` on execution requests.
+5. Treat invalid SemVer `400` as a client defect.
+6. Treat stale-service `412` as non-retryable until the daemon is updated or
+   a verified compatible daemon is selected.
+
+The precondition is checked before handler state creation and is separate from
+bearer authentication.

package/.aiwg/addons/omnius-docs/skills/runtime-provenance-audit/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: runtime-provenance-audit
+description: Prove active process, artifact, port, configuration, and restart ownership for runtime claims.
 compaction_strategy: structured
 ---
 
@@ -17,4 +19,3 @@ port, endpoint, or update.
    and whether a process is project-supervised or ad hoc.
 4. Report unavailable access as `blocked` or `unverified`, never as a failed
    implementation claim.
-

package/.aiwg/addons/omnius-docs/skills/secrets-and-config-audit/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: secrets-and-config-audit
+description: Audit credential and configuration provenance without exposing secret values.
 compaction_strategy: structured
 ---
 
@@ -19,4 +21,3 @@ configuration provenance.
    source indicator, timestamp, and non-secret receipt only.
 5. Treat 401/403, unavailable models, or changed free/pricing status as a
    configuration outcome requiring action—not an invitation to retry blindly.
-

package/.aiwg/addons/omnius-docs/skills/test-surface-audit/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: test-surface-audit
+description: Map canonical tests to claims and expose placeholder, mock-only, copied-logic, and unwired verification gaps.
 compaction_strategy: structured
 ---
 
@@ -19,4 +21,3 @@ Use when a task changes behavior, claims a fix, or asks for release readiness.
    passing execution from coverage of the requested behavior.
 5. On failure, retain the exact diagnostic and create a distinct recovery
    action; do not blanket-retry.
-

package/.aiwg/addons/omnius-docs/skills/workspace-reality-audit/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: workspace-reality-audit
+description: Audit repository truth, generated artifacts, documentation drift, and delivery ownership before claiming progress.
 compaction_strategy: structured
 ---
 
@@ -21,4 +23,3 @@ completion claim.
 
 For each finding emit: `status`, `evidence`, `impact`, `counter-check`, and
 `smallest next action`. Keep workspace findings separate from runtime claims.
-

package/.aiwg/addons/omnius-rest-docs/README.md

@@ -5,3 +5,6 @@ This project-local AIWG-shaped addon exposes the Omnius REST API reference throu
 Agents should discover it with `skill_list filter="rest"` or `/skills rest`, then load `omnius-rest-docs` only when the current task needs REST API guidance.
 
 The human-readable docs live in `docs/rest/`. The machine-readable contract is generated by `packages/cli/src/api/openapi.ts` and served by the daemon at `/openapi.json`.
+
+For intent-first lookup, use `omnius discover`, `GET /v1/discovery`, or the
+bundled `docs/DISCOVERY.json` before opening a REST family page.

package/.aiwg/addons/omnius-rest-docs/manifest.json

@@ -1,24 +1,31 @@
 {
-  "name": "omnius-rest-docs",
-  "version": "1.0.0",
+  "id": "omnius-rest-docs",
+  "type": "addon",
+  "name": "Omnius REST Documentation",
+  "version": "1.1.0",
   "description": "Project-local Omnius REST API documentation bundle for discover-then-expand agent workflows.",
-  "skills": [
-    {
-      "name": "omnius-rest-docs",
-      "description": "Explore Omnius REST API docs, OpenAPI source, auth scopes, realtime chat, tools, memory, AIWG, voice, sponsor, and governance endpoints.",
-      "triggers": [
-        "Omnius REST API",
-        "/v1 endpoints",
-        "OpenAPI",
-        "Swagger UI",
-        "/openapi.json",
-        "API key scopes",
-        "realtime REST",
-        "memory REST endpoint",
-        "skills REST endpoint",
-        "AIWG cascade endpoint",
-        "voicechat websocket"
-      ]
+  "manifestVersion": "1",
+  "platforms": {
+    "claude": "full",
+    "codex": "full",
+    "factory": "full",
+    "cursor": "full",
+    "generic": "full"
+  },
+  "keywords": [
+    "omnius",
+    "rest",
+    "openapi",
+    "discovery",
+    "tools",
+    "project-local"
+  ],
+  "deployment": {
+    "pathTemplate": ".{platform}/skills/{id}.md"
+  },
+  "addonConfig": {
+    "entry": {
+      "skills": "skills/"
     }
-  ]
+  }
 }

package/.aiwg/addons/omnius-rest-docs/skills/omnius-rest-docs/SKILL.md

@@ -1,4 +1,6 @@
 ---
+name: omnius-rest-docs
+description: Route Omnius REST questions to the smallest endpoint-family guide, OpenAPI contract, auth reference, or example.
 compaction_strategy: structured
 ---
 
@@ -28,15 +30,17 @@ Explore Omnius REST API documentation without loading the whole API reference in
 ## Workflow
 
 1. Start with `docs/rest/INDEX.md`.
-2. Identify the endpoint family that matches the user's question.
-3. Open `docs/reference/rest-api.md` when the user needs endpoint inventory or method/path details.
-4. Open only the relevant file under `docs/rest/endpoints/` or `docs/rest/examples/` for family-specific usage.
-5. Treat `packages/cli/src/api/openapi.ts` and the live daemon `/openapi.json` as the canonical machine contract.
-6. Prefer concise endpoint-specific guidance over copying long tables.
+2. For intent lookup, search `docs/DISCOVERY.json` or `GET /v1/discovery`.
+3. Identify the endpoint family that matches the user's question.
+4. Open `docs/reference/rest-api.md` when the user needs endpoint inventory or method/path details.
+5. Open only the relevant file under `docs/rest/endpoints/` or `docs/rest/examples/` for family-specific usage.
+6. Treat `packages/cli/src/api/openapi.ts` and the live daemon `/openapi.json` as the canonical machine contract.
+7. Prefer concise endpoint-specific guidance over copying long tables.
 
 ## Endpoint Family Map
 
 - Chat, realtime, OpenAI-compatible inference: `docs/rest/endpoints/chat.md`
+- Capability discovery: `docs/rest/endpoints/discovery.md`
 - Full endpoint inventory: `docs/reference/rest-api.md`
 - Agent runs, jobs, todos, scheduled tasks: `docs/rest/endpoints/run.md`
 - Config, keys, profiles, projects: `docs/rest/endpoints/config.md`

package/README.md

@@ -30,6 +30,31 @@ omnius serve
 
 The daemon defaults to `http://127.0.0.1:11435`. Open the interactive API docs at `http://127.0.0.1:11435/docs`.
 
+## Agent Discovery
+
+The npm package ships its complete documentation and a machine-readable
+capability catalog. An agent does not need to inspect Omnius source or guess
+which endpoint owns a capability:
+
+```bash
+omnius discover "bring your own inference"
+omnius show provider.anthropic
+omnius show provider.gemini
+omnius show tool.web-search
+omnius capabilities --json
+```
+
+With the daemon running, the same discovery cascade is available at
+`GET /v1/discovery`, with exact entry expansion at
+`GET /v1/discovery/{id}`. The live API contract remains available at
+`/openapi.json`, direct tool metadata at `/v1/tools`, and skills at
+`/v1/skills`.
+
+Start with [the discovery guide](docs/DISCOVERY.md) when integrating another
+agent or service. Use [bring-your-own inference](docs/guides/bring-your-own-inference.md)
+for provider protocols and keys, and [tools and web search](docs/guides/tools-and-web-search.md)
+for the distinction between direct tools and agent-bound tools.
+
 ## What Omnius Does
 
 - Runs autonomous coding tasks, edits files, executes tools, tests changes, and iterates on failures.
@@ -67,6 +92,9 @@ omnius serve
 | Area | What to read |
 | --- | --- |
 | Install and setup | [Install](docs/getting-started/install.md), [First run](docs/getting-started/first-run.md), [Model providers](docs/getting-started/model-providers.md) |
+| Agent discovery | [Discovery cascade](docs/DISCOVERY.md), [machine catalog](docs/DISCOVERY.json), [agent integration](docs/guides/agent-integration.md) |
+| Bring your own inference | [Provider protocols and keys](docs/guides/bring-your-own-inference.md) |
+| Tools and web search | [Tool discovery and invocation](docs/guides/tools-and-web-search.md) |
 | Terminal workflows | [TUI workflows](docs/guides/tui-workflows.md), [Slash commands](docs/reference/slash-commands.md) |
 | REST daemon | [REST reference](docs/reference/rest-api.md), [REST quickref](docs/rest/QUICKREF.md), [OpenAPI source](docs/rest/openapi-source.md) |
 | Realtime voice chat | [Realtime guide](docs/guides/realtime.md) |
@@ -74,6 +102,7 @@ omnius serve
 | Telegram bridge | [Telegram guide](docs/guides/telegram.md) |
 | Media generation | [Media guide](docs/guides/media-generation.md) |
 | Operations | [Runtime hygiene](docs/operations/runtime-hygiene.md), [Security and remote access](docs/operations/security-and-remote-access.md) |
+| Service compatibility | [Runtime version gate](docs/operations/version-compatibility.md) |
 | Architecture | [Architecture overview](docs/architecture/overview.md) |
 | Agent-explorable docs | [Agent memory docs index](docs/agent-memory/INDEX.md) |
 
@@ -173,6 +202,13 @@ The complete endpoint inventory follows. It is kept in lockstep with the served
 | `GET` | `/v1/cost` | Cost tracker |
 | `GET` | `/v1/system` | CPU, RAM, GPU, and system snapshot |
 
+### Discovery
+
+| Method | Path | Purpose |
+| --- | --- | --- |
+| `GET` | `/v1/discovery` | Search or list the bundled capability catalog |
+| `GET` | `/v1/discovery/{id}` | Expand one stable capability entry |
+
 ### Inference and chat
 
 | Method | Path | Purpose |

package/dist/discovery.d.ts

@@ -0,0 +1,50 @@
+export type DiscoveryKind = "capability" | "provider" | "tool" | "api" | "command" | "skill" | "guide" | "config" | "operation";
+export interface DiscoveryInterface {
+    type: string;
+    target: string;
+    description?: string;
+}
+export interface DiscoveryReference {
+    type: string;
+    target: string;
+    relation?: string;
+}
+export interface DiscoveryEntry {
+    id: string;
+    kind: DiscoveryKind;
+    title: string;
+    summary: string;
+    aliases?: string[];
+    keywords?: string[];
+    maturity?: string;
+    prerequisites?: string[];
+    interfaces?: DiscoveryInterface[];
+    references?: DiscoveryReference[];
+    related?: string[];
+    [key: string]: unknown;
+}
+export interface DiscoveryCatalog {
+    schema_version: string;
+    generated_at?: string;
+    package_version?: string;
+    entries: DiscoveryEntry[];
+    [key: string]: unknown;
+}
+export interface DiscoveryQueryOptions {
+    kind?: DiscoveryKind;
+    limit?: number;
+    offset?: number;
+}
+export interface DiscoveryResult {
+    score: number;
+    entry: DiscoveryEntry;
+}
+export declare function findDiscoveryCatalogPath(): string | null;
+export declare function loadDiscoveryCatalog(): DiscoveryCatalog;
+export declare function discoverCapabilities(query?: string, options?: DiscoveryQueryOptions, catalog?: DiscoveryCatalog): DiscoveryResult[];
+export declare function rankCapabilities(query?: string, kind?: DiscoveryKind, catalog?: DiscoveryCatalog): DiscoveryResult[];
+export declare function showCapability(id: string, catalog?: DiscoveryCatalog): DiscoveryEntry | undefined;
+export declare function renderDiscoveryEntry(entry: DiscoveryEntry): string;
+export declare function discoverySchemaVersion(): string;
+export declare function discoveryDocsRoot(): string | null;
+//# sourceMappingURL=discovery.d.ts.map