@cyanheads/nist-nvd-mcp-server 0.2.0 → 0.3.1

package/AGENTS.md

@@ -1,11 +1,11 @@
 # Developer Protocol
 
 **Server:** nist-nvd-mcp-server
-**Version:** 0.2.0
-**Framework:** [@cyanheads/mcp-ts-core](https://www.npmjs.com/package/@cyanheads/mcp-ts-core) `^0.11.0`
-**Engines:** Bun ≥1.3.0, Node ≥24.0.0
-**MCP SDK:** `@modelcontextprotocol/sdk` ^1.29.0
-**Zod:** ^4.4.3
+**Version:** 0.3.1
+**Framework:** [@cyanheads/mcp-ts-core](https://www.npmjs.com/package/@cyanheads/mcp-ts-core) `^0.13.6`
+**Engines:** Bun ≥1.4.0, Node ≥24.0.0
+**MCP SDK:** `@modelcontextprotocol/server` ^2.0.0
+**Zod:** ^4.6.5
 
 > **Read the framework docs first:** `node_modules/@cyanheads/mcp-ts-core/CLAUDE.md` contains the full API reference — builders, Context, error codes, exports, patterns. This file covers server-specific conventions only.
 
@@ -35,7 +35,7 @@ Tailor suggestions to what's actually missing or stale — don't recite the full
 - **Logic throws, framework catches.** Tool/resource handlers are pure — throw on failure, no `try/catch`. Plain `Error` is fine; the framework catches, classifies, and formats. Use error factories (`notFound()`, `validationError()`, etc.) when the error code matters.
 - **Use `ctx.log`** for request-scoped logging. No `console` calls.
 - **Use `ctx.state`** for tenant-scoped storage. Never access persistence directly.
-- **Check `ctx.elicit`** for presence before calling.
+- **Need input the caller didn't supply?** `return ctx.requestInput(...)` and read `ctx.inputs` when the handler is re-entered. Never `await` for user input mid-handler.
 - **Secrets in env vars only** — never hardcoded.
 - **Close the loop on issues.** When implementing work tracked by a GitHub issue, comment on the issue with what landed and close it. Do both — a comment without a close leaves stale issues open; a close without a comment leaves no record of what shipped. The comment is for future readers — state the concrete changes, not the conversation that produced them.
 
@@ -158,6 +158,23 @@ export function getServerConfig() {
 
 `parseEnvConfig` maps Zod schema paths → env var names so errors name the variable (`MY_API_KEY`) not the path (`apiKey`). Throws `ConfigurationError`, which the framework prints as a clean startup banner.
 
+For env booleans use `z.stringbool()`, never `z.coerce.boolean()` — `Boolean("false")` is `true`, so a coerced flag can't be disabled through the environment. `z.stringbool()` parses `true/false/1/0/yes/no/on/off` and rejects anything else, so `=false` actually disables.
+
+### Server identity and instructions
+
+`createApp()` accepts optional identity fields forwarded to the SDK's `initialize` response and the server manifest (`/.well-known/mcp.json`):
+
+```ts
+await createApp({
+  name: 'nist-nvd-mcp-server',
+  title: 'nist-nvd-mcp-server',                 // human-readable display name
+  websiteUrl: 'https://github.com/cyanheads/nist-nvd-mcp-server',
+  instructions: 'Use nvd_search_cpes to resolve a product to its CPE name before auditing it.',
+});
+```
+
+`instructions` is optional server-level orientation, sent on every `initialize` as session-level context. Use it for deployment guidance (connection aliases, regional notes, scope hints) instead of repeating the same context across tool descriptions. Client adoption is uneven, but there's no downside when set.
+
 ---
 
 ## Context
@@ -166,13 +183,13 @@ Handlers receive a unified `ctx` object. Key properties:
 
 | Property | Description |
 |:---------|:------------|
-| `ctx.log` | Request-scoped logger — `.debug()`, `.info()`, `.notice()`, `.warning()`, `.error()`. Auto-correlates requestId, traceId, tenantId. |
+| `ctx.log` | Request-scoped logger — `.debug()`, `.info()`, `.notice()`, `.warning()`, `.error()`. Auto-correlates requestId, traceId, tenantId. Dual-sink: Pino **and** `notifications/message` to the client, so treat it as client-visible. |
 | `ctx.state` | Tenant-scoped KV — `.get(key)`, `.set(key, value, { ttl? })`, `.delete(key)`, `.getMany(keys)`, `.list(prefix, { cursor, limit })`. Accepts any serializable value. |
-| `ctx.elicit` | Ask user for structured input — form call `(message, schema)` or `.url(message, url)` for an external link. **Check for presence first:** `if (ctx.elicit) { ... }` |
+| `ctx.requestInput` | Suspend and ask the caller for more input — `return ctx.requestInput({ inputRequests: { key: inputRequired.elicit({ message, requestedSchema }) } })`. Never returns; the handler is re-entered with the answers. Always present. |
+| `ctx.inputs` | Reader over a retried request's responses — `.accepted(key, schema)`, `.view(key)`, `.state()`, `.dropped`. Empty on the first round. |
 | `ctx.enrich` | Success-path agent context (empty-result notices, query echo, pagination totals) — `ctx.enrich(...)` or `.notice()` / `.total()` / `.echo()` / `.truncated()`. Reaches `structuredContent` and `content[]`; lands only when the definition declares an `enrichment` block (no-op otherwise). |
 | `ctx.content` | Non-text content blocks — `.image(data, mimeType)`, `.audio(data, mimeType)`, or `ctx.content(block)` for a raw block. Prepended to `content[]` after `format()`; never enters `structuredContent`. |
 | `ctx.signal` | `AbortSignal` for cancellation. Forwarded to NVD HTTP requests. |
-| `ctx.progress` | Task progress (present when `task: true`) — `.setTotal(n)`, `.increment()`, `.update(message)`. |
 | `ctx.requestId` | Unique request ID. |
 | `ctx.tenantId` | Tenant ID from JWT; `'default'` for stdio or HTTP with auth off. |
 
@@ -182,7 +199,7 @@ Handlers receive a unified `ctx` object. Key properties:
 
 Handlers throw — the framework catches, classifies, and formats.
 
-**Recommended: typed error contract.** Declare `errors: [{ reason, code, when, recovery, retryable? }]` on `tool()` / `resource()` to receive `ctx.fail(reason, …)` typed against the reason union. TypeScript catches typos at compile time, `data.reason` is auto-populated for observability, linter enforces conformance against the handler body. `recovery` is required descriptive metadata for the agent's next move (≥ 5 words, lint-validated); for the wire `data.recovery.hint` (mirrored into `content[]` text), pass explicitly at the throw site when dynamic context matters: `ctx.fail('reason', msg, { recovery: { hint: '...' } })`. Baseline codes (`InternalError`, `ServiceUnavailable`, `Timeout`, `ValidationError`, `SerializationError`) bubble freely and don't need declaring.
+**Recommended: typed error contract.** Declare `errors: [{ reason, code, when, recovery, retryable?, severity?, thrownBy? }]` on `tool()` / `resource()` to receive `ctx.fail(reason, …)` typed against the reason union. TypeScript catches typos at compile time, `data.reason` is auto-populated for observability, linter enforces conformance against the handler body. `recovery` is required (≥ 5 words, lint-validated) — the single source of truth for the agent's next move. Pass `ctx.recoveryFor('reason')` as the throw's data to put it on the wire (`data.recovery.hint`, mirrored into `content[]` text unless the message already contains it verbatim); override with an explicit `{ recovery: { hint: '...' } }` when dynamic runtime context matters. Forwarding it is lint-enforced per throw site (`error-contract-recovery-unforwarded`). Mark an entry the service layer throws with `thrownBy: 'service'` so `error-contract-unthrown` skips it — lint-only metadata, nothing at runtime reads it. Baseline codes (`InternalError`, `ServiceUnavailable`, `Timeout`, `ValidationError`, `SerializationError`, `RequestCancelled`) bubble freely and don't need declaring.
 
 ```ts
 import { JsonRpcErrorCode } from '@cyanheads/mcp-ts-core/errors';
@@ -194,7 +211,7 @@ errors: [
 ],
 async handler(input, ctx) {
   const item = await db.find(input.id);
-  if (!item) throw ctx.fail('no_match', `No item ${input.id}`);
+  if (!item) throw ctx.fail('no_match', `No item ${input.id}`, ctx.recoveryFor('no_match'));
   return item;
 }
 ```
@@ -272,9 +289,9 @@ src/
 
 ## Skills
 
-Skills are modular instructions in `skills/` at the project root. Read them directly when a task matches — e.g., `skills/add-tool/SKILL.md` when adding a tool.
+Skills are modular instructions in `framework-skills/` at the project root. Read them directly when a task matches — e.g., `framework-skills/add-tool/SKILL.md` when adding a tool. `bun run list-skills` prints the full registry. The directory is deliberately not `skills/`: Claude Code and Codex auto-load a plugin's root `skills/`, so a server that ships `.claude-plugin/` or `.codex-plugin/` would hand these development skills to every agent that installs it. Keep `skills/` free for skills meant for those agents.
 
-**Agent skill directory:** Copy skills into the directory your agent discovers (Claude Code: `.claude/skills/`, others: equivalent). Skills then load as context without referencing `skills/` paths. After framework updates, run the `maintenance` skill — Phase B re-syncs the agent directory.
+**Agent skill directory:** Copy skills into the directory your agent discovers (Claude Code: `.claude/skills/`, others: equivalent). Skills then load as context without referencing `framework-skills/` paths. After framework updates, run the `maintenance` skill — Phase B re-syncs the agent directory.
 
 Available skills:
 
@@ -293,8 +310,9 @@ Available skills:
 | `security-pass` | Audit server for MCP-flavored security gaps: output injection, scope blast radius, input sinks, tenant isolation |
 | `code-simplifier` | Post-session cleanup against `git diff` — modernize syntax, consolidate duplication, align with the codebase |
 | `polish-docs-meta` | Finalize docs, README, metadata, and agent protocol for shipping |
-| `git-wrapup` | Land working-tree changes as a versioned commit + annotated tag — version bump, changelog, verify, tag. Local only. |
-| `release-and-publish` | Push + npm + MCP Registry + GH Release + Docker. Picks up from `git-wrapup` |
+| `git-wrapup` | Land working-tree changes as a commit stack — version bump, changelog, verify, commit by concern, release commit on top. No tag, no push to main; opens the release PR when the project declares release PR mode |
+| `release-pr-review` | Review pass on an open release PR — simplifier + correctness review, fixes as ordinary commits on top of the stack, PR body kept in sync. Release PR mode only |
+| `release-and-publish` | Fast-forward merge (release PR mode) + tag + push + npm + MCP Registry + GH Release + Docker. Picks up from `git-wrapup` |
 | `maintenance` | Investigate changelogs, adopt upstream changes, sync skills to agent dirs |
 | `orchestrations` | Chain task skills into a gated multi-phase pipeline — build-out, QA-fix, update-ship — when you can spawn sub-agents |
 | `report-issue-framework` | File a bug or feature request against `@cyanheads/mcp-ts-core` via `gh` CLI |
@@ -303,7 +321,7 @@ Available skills:
 | `api-auth` | Auth modes, scopes, JWT/OAuth |
 | `api-canvas` | DataCanvas: register tabular data, run SQL, export, plus the `spillover()` helper for big result sets — Tier 3 opt-in |
 | `api-config` | AppConfig, parseConfig, env vars |
-| `api-context` | Context interface, logger, state, progress |
+| `api-context` | Context interface, RequestContext, logger, state, multi-round-trip input |
 | `api-errors` | McpError, JsonRpcErrorCode, error patterns |
 | `api-linter` | Definition linter rule catalog — invoked by `bun run lint:mcp` and `devcheck` |
 | `api-mirror` | MirrorService: persistent self-refreshing local mirror (embedded SQLite + FTS5) of a bulk upstream dataset — Tier 3 opt-in |
@@ -346,11 +364,11 @@ When you complete a skill's checklist, check the boxes and add a completion time
 
 ## Bundling
 
-`npm run bundle` produces a `.mcpb` extension bundle for one-click install in Claude Desktop. MCPB is stdio-only — HTTP and Cloudflare Workers deployments are unaffected. Consumers who don't need it can delete `manifest.json` and `.mcpbignore`; `lint:packaging` skips cleanly.
+`npm run bundle` produces a `.mcpb` extension bundle for one-click install in Claude Desktop. The pack step is followed by `scripts/clean-mcpb.ts`, which prunes dev dependencies (`mcpb clean`) and strips two classes of `node_modules/**` content that root-anchored `.mcpbignore` patterns cannot reach: dependency-shipped agent docs (`framework-skills/`, `skills/`, `.claude/`, `.agents/`, `SKILL.md`) and platform-specific native bindings, which would otherwise lock the bundle to the platform it was packed on. MCPB is stdio-only — HTTP and Cloudflare Workers deployments are unaffected. Consumers who don't need it can delete `manifest.json` and `.mcpbignore`; `lint:packaging` skips cleanly.
 
 **Adding an env var requires both files:** `server.json` (registry discovery, `environmentVariables[]`) and `manifest.json` (bundle install UX, `mcp_config.env` + `user_config`). `lint:packaging` (run by `devcheck`) verifies the env var names match.
 
-**README install badges** (Claude Desktop `.mcpb`, Cursor, VS Code) and the `base64` / `encodeURIComponent` config-generation commands are ship-time concerns — run the `polish-docs-meta` skill, which carries the badge format, layout, and generation snippets in `skills/polish-docs-meta/references/readme.md`.
+**README install badges** (Claude Desktop `.mcpb`, Cursor, VS Code) and the `base64` / `encodeURIComponent` config-generation commands are ship-time concerns — run the `polish-docs-meta` skill, which carries the badge format, layout, and generation snippets in `framework-skills/polish-docs-meta/references/readme.md`.
 
 ---
 
@@ -364,23 +382,29 @@ Each per-version file opens with YAML frontmatter:
 ---
 summary: "One-line headline, ≤350 chars"  # required — powers the rollup index
 breaking: false                            # optional — true flags breaking changes
-security: false                            # optional — true flags security fixes
+security: false                            # optional — true ONLY for a source-code security fix, never a dependency CVE bump
 ---
 
 # 0.1.0 — YYYY-MM-DD
 ...
 ```
 
-`breaking: true` renders a `· ⚠️ Breaking` badge — use it when consumers must update code on upgrade (signature changes, removed APIs, config renames). `security: true` renders a `· 🛡️ Security` badge and pairs with a `## Security` body section. When both are set, badges render `· ⚠️ Breaking · 🛡️ Security`.
+`breaking: true` renders a `· ⚠️ Breaking` badge — use it when consumers must update code on upgrade (signature changes, removed APIs, config renames). `security: true` renders a `· 🛡️ Security` badge and pairs with a `## Security` body section — set it only for a security fix in this server's *own source code*, never for a routine dependency or transitive CVE bump (record those under `## Dependencies`). When both are set, badges render `· ⚠️ Breaking · 🛡️ Security`.
 
 `agent-notes` is an optional free-form field for maintenance agents processing the release downstream. Content here won't appear in the rendered CHANGELOG — it's consumed by agents running the `maintenance` skill. Use it for adoption instructions that don't fit the human-facing sections: new files to create, fields to populate, one-time migration steps. Omit entirely when there's nothing to say.
 
-**Section order** (Keep a Changelog): Added, Changed, Deprecated, Removed, Fixed, Security. Include only sections with entries — don't ship empty headers.
+**Section order:** the Keep a Changelog sequence — Added, Changed, Deprecated, Removed, Fixed, Security — then `Dependencies` last. Include only sections with entries — don't ship empty headers.
 
 **Tag annotations** render as GitHub Release bodies via `--notes-from-tag`. They must be structured markdown — never a flat comma-separated string. Subject omits the version number (GitHub prepends it). See `changelog/template.md` for the full format reference.
 
 ---
 
+## Publishing
+
+**Every release goes through a release PR, straight-through** — `git-wrapup`'s "Release PR mode", mode `straight-through`. One run: `git-wrapup` lands the commit stack on `release/<version>`, pushes it, and opens the PR (title = the release commit subject, body = the changelog entry plus a gates section); `release-and-publish` then fast-forwards `main` locally with `git merge --ff-only`, creates the tag on `main`'s tip, pushes `main` and the tag, deletes the branch, and publishes. A caller's brief may run a given release as `gated` instead — a `release-pr-review` pass on the open PR before `release-and-publish`. **Never merge through the GitHub UI or `gh pr merge`**: squash and rebase-merge are disabled in the repo settings because both rewrite the stack (rebase-merge also strips the SSH signatures), and a merge commit breaks the linear history.
+
+---
+
 ## Imports
 
 ```ts

package/CLAUDE.md

@@ -1,11 +1,11 @@
 # Developer Protocol
 
 **Server:** nist-nvd-mcp-server
-**Version:** 0.2.0
-**Framework:** [@cyanheads/mcp-ts-core](https://www.npmjs.com/package/@cyanheads/mcp-ts-core) `^0.11.0`
-**Engines:** Bun ≥1.3.0, Node ≥24.0.0
-**MCP SDK:** `@modelcontextprotocol/sdk` ^1.29.0
-**Zod:** ^4.4.3
+**Version:** 0.3.1
+**Framework:** [@cyanheads/mcp-ts-core](https://www.npmjs.com/package/@cyanheads/mcp-ts-core) `^0.13.6`
+**Engines:** Bun ≥1.4.0, Node ≥24.0.0
+**MCP SDK:** `@modelcontextprotocol/server` ^2.0.0
+**Zod:** ^4.6.5
 
 > **Read the framework docs first:** `node_modules/@cyanheads/mcp-ts-core/CLAUDE.md` contains the full API reference — builders, Context, error codes, exports, patterns. This file covers server-specific conventions only.
 
@@ -35,7 +35,7 @@ Tailor suggestions to what's actually missing or stale — don't recite the full
 - **Logic throws, framework catches.** Tool/resource handlers are pure — throw on failure, no `try/catch`. Plain `Error` is fine; the framework catches, classifies, and formats. Use error factories (`notFound()`, `validationError()`, etc.) when the error code matters.
 - **Use `ctx.log`** for request-scoped logging. No `console` calls.
 - **Use `ctx.state`** for tenant-scoped storage. Never access persistence directly.
-- **Check `ctx.elicit`** for presence before calling.
+- **Need input the caller didn't supply?** `return ctx.requestInput(...)` and read `ctx.inputs` when the handler is re-entered. Never `await` for user input mid-handler.
 - **Secrets in env vars only** — never hardcoded.
 - **Close the loop on issues.** When implementing work tracked by a GitHub issue, comment on the issue with what landed and close it. Do both — a comment without a close leaves stale issues open; a close without a comment leaves no record of what shipped. The comment is for future readers — state the concrete changes, not the conversation that produced them.
 
@@ -158,6 +158,23 @@ export function getServerConfig() {
 
 `parseEnvConfig` maps Zod schema paths → env var names so errors name the variable (`MY_API_KEY`) not the path (`apiKey`). Throws `ConfigurationError`, which the framework prints as a clean startup banner.
 
+For env booleans use `z.stringbool()`, never `z.coerce.boolean()` — `Boolean("false")` is `true`, so a coerced flag can't be disabled through the environment. `z.stringbool()` parses `true/false/1/0/yes/no/on/off` and rejects anything else, so `=false` actually disables.
+
+### Server identity and instructions
+
+`createApp()` accepts optional identity fields forwarded to the SDK's `initialize` response and the server manifest (`/.well-known/mcp.json`):
+
+```ts
+await createApp({
+  name: 'nist-nvd-mcp-server',
+  title: 'nist-nvd-mcp-server',                 // human-readable display name
+  websiteUrl: 'https://github.com/cyanheads/nist-nvd-mcp-server',
+  instructions: 'Use nvd_search_cpes to resolve a product to its CPE name before auditing it.',
+});
+```
+
+`instructions` is optional server-level orientation, sent on every `initialize` as session-level context. Use it for deployment guidance (connection aliases, regional notes, scope hints) instead of repeating the same context across tool descriptions. Client adoption is uneven, but there's no downside when set.
+
 ---
 
 ## Context
@@ -166,13 +183,13 @@ Handlers receive a unified `ctx` object. Key properties:
 
 | Property | Description |
 |:---------|:------------|
-| `ctx.log` | Request-scoped logger — `.debug()`, `.info()`, `.notice()`, `.warning()`, `.error()`. Auto-correlates requestId, traceId, tenantId. |
+| `ctx.log` | Request-scoped logger — `.debug()`, `.info()`, `.notice()`, `.warning()`, `.error()`. Auto-correlates requestId, traceId, tenantId. Dual-sink: Pino **and** `notifications/message` to the client, so treat it as client-visible. |
 | `ctx.state` | Tenant-scoped KV — `.get(key)`, `.set(key, value, { ttl? })`, `.delete(key)`, `.getMany(keys)`, `.list(prefix, { cursor, limit })`. Accepts any serializable value. |
-| `ctx.elicit` | Ask user for structured input — form call `(message, schema)` or `.url(message, url)` for an external link. **Check for presence first:** `if (ctx.elicit) { ... }` |
+| `ctx.requestInput` | Suspend and ask the caller for more input — `return ctx.requestInput({ inputRequests: { key: inputRequired.elicit({ message, requestedSchema }) } })`. Never returns; the handler is re-entered with the answers. Always present. |
+| `ctx.inputs` | Reader over a retried request's responses — `.accepted(key, schema)`, `.view(key)`, `.state()`, `.dropped`. Empty on the first round. |
 | `ctx.enrich` | Success-path agent context (empty-result notices, query echo, pagination totals) — `ctx.enrich(...)` or `.notice()` / `.total()` / `.echo()` / `.truncated()`. Reaches `structuredContent` and `content[]`; lands only when the definition declares an `enrichment` block (no-op otherwise). |
 | `ctx.content` | Non-text content blocks — `.image(data, mimeType)`, `.audio(data, mimeType)`, or `ctx.content(block)` for a raw block. Prepended to `content[]` after `format()`; never enters `structuredContent`. |
 | `ctx.signal` | `AbortSignal` for cancellation. Forwarded to NVD HTTP requests. |
-| `ctx.progress` | Task progress (present when `task: true`) — `.setTotal(n)`, `.increment()`, `.update(message)`. |
 | `ctx.requestId` | Unique request ID. |
 | `ctx.tenantId` | Tenant ID from JWT; `'default'` for stdio or HTTP with auth off. |
 
@@ -182,7 +199,7 @@ Handlers receive a unified `ctx` object. Key properties:
 
 Handlers throw — the framework catches, classifies, and formats.
 
-**Recommended: typed error contract.** Declare `errors: [{ reason, code, when, recovery, retryable? }]` on `tool()` / `resource()` to receive `ctx.fail(reason, …)` typed against the reason union. TypeScript catches typos at compile time, `data.reason` is auto-populated for observability, linter enforces conformance against the handler body. `recovery` is required descriptive metadata for the agent's next move (≥ 5 words, lint-validated); for the wire `data.recovery.hint` (mirrored into `content[]` text), pass explicitly at the throw site when dynamic context matters: `ctx.fail('reason', msg, { recovery: { hint: '...' } })`. Baseline codes (`InternalError`, `ServiceUnavailable`, `Timeout`, `ValidationError`, `SerializationError`) bubble freely and don't need declaring.
+**Recommended: typed error contract.** Declare `errors: [{ reason, code, when, recovery, retryable?, severity?, thrownBy? }]` on `tool()` / `resource()` to receive `ctx.fail(reason, …)` typed against the reason union. TypeScript catches typos at compile time, `data.reason` is auto-populated for observability, linter enforces conformance against the handler body. `recovery` is required (≥ 5 words, lint-validated) — the single source of truth for the agent's next move. Pass `ctx.recoveryFor('reason')` as the throw's data to put it on the wire (`data.recovery.hint`, mirrored into `content[]` text unless the message already contains it verbatim); override with an explicit `{ recovery: { hint: '...' } }` when dynamic runtime context matters. Forwarding it is lint-enforced per throw site (`error-contract-recovery-unforwarded`). Mark an entry the service layer throws with `thrownBy: 'service'` so `error-contract-unthrown` skips it — lint-only metadata, nothing at runtime reads it. Baseline codes (`InternalError`, `ServiceUnavailable`, `Timeout`, `ValidationError`, `SerializationError`, `RequestCancelled`) bubble freely and don't need declaring.
 
 ```ts
 import { JsonRpcErrorCode } from '@cyanheads/mcp-ts-core/errors';
@@ -194,7 +211,7 @@ errors: [
 ],
 async handler(input, ctx) {
   const item = await db.find(input.id);
-  if (!item) throw ctx.fail('no_match', `No item ${input.id}`);
+  if (!item) throw ctx.fail('no_match', `No item ${input.id}`, ctx.recoveryFor('no_match'));
   return item;
 }
 ```
@@ -272,9 +289,9 @@ src/
 
 ## Skills
 
-Skills are modular instructions in `skills/` at the project root. Read them directly when a task matches — e.g., `skills/add-tool/SKILL.md` when adding a tool.
+Skills are modular instructions in `framework-skills/` at the project root. Read them directly when a task matches — e.g., `framework-skills/add-tool/SKILL.md` when adding a tool. `bun run list-skills` prints the full registry. The directory is deliberately not `skills/`: Claude Code and Codex auto-load a plugin's root `skills/`, so a server that ships `.claude-plugin/` or `.codex-plugin/` would hand these development skills to every agent that installs it. Keep `skills/` free for skills meant for those agents.
 
-**Agent skill directory:** Copy skills into the directory your agent discovers (Claude Code: `.claude/skills/`, others: equivalent). Skills then load as context without referencing `skills/` paths. After framework updates, run the `maintenance` skill — Phase B re-syncs the agent directory.
+**Agent skill directory:** Copy skills into the directory your agent discovers (Claude Code: `.claude/skills/`, others: equivalent). Skills then load as context without referencing `framework-skills/` paths. After framework updates, run the `maintenance` skill — Phase B re-syncs the agent directory.
 
 Available skills:
 
@@ -293,8 +310,9 @@ Available skills:
 | `security-pass` | Audit server for MCP-flavored security gaps: output injection, scope blast radius, input sinks, tenant isolation |
 | `code-simplifier` | Post-session cleanup against `git diff` — modernize syntax, consolidate duplication, align with the codebase |
 | `polish-docs-meta` | Finalize docs, README, metadata, and agent protocol for shipping |
-| `git-wrapup` | Land working-tree changes as a versioned commit + annotated tag — version bump, changelog, verify, tag. Local only. |
-| `release-and-publish` | Push + npm + MCP Registry + GH Release + Docker. Picks up from `git-wrapup` |
+| `git-wrapup` | Land working-tree changes as a commit stack — version bump, changelog, verify, commit by concern, release commit on top. No tag, no push to main; opens the release PR when the project declares release PR mode |
+| `release-pr-review` | Review pass on an open release PR — simplifier + correctness review, fixes as ordinary commits on top of the stack, PR body kept in sync. Release PR mode only |
+| `release-and-publish` | Fast-forward merge (release PR mode) + tag + push + npm + MCP Registry + GH Release + Docker. Picks up from `git-wrapup` |
 | `maintenance` | Investigate changelogs, adopt upstream changes, sync skills to agent dirs |
 | `orchestrations` | Chain task skills into a gated multi-phase pipeline — build-out, QA-fix, update-ship — when you can spawn sub-agents |
 | `report-issue-framework` | File a bug or feature request against `@cyanheads/mcp-ts-core` via `gh` CLI |
@@ -303,7 +321,7 @@ Available skills:
 | `api-auth` | Auth modes, scopes, JWT/OAuth |
 | `api-canvas` | DataCanvas: register tabular data, run SQL, export, plus the `spillover()` helper for big result sets — Tier 3 opt-in |
 | `api-config` | AppConfig, parseConfig, env vars |
-| `api-context` | Context interface, logger, state, progress |
+| `api-context` | Context interface, RequestContext, logger, state, multi-round-trip input |
 | `api-errors` | McpError, JsonRpcErrorCode, error patterns |
 | `api-linter` | Definition linter rule catalog — invoked by `bun run lint:mcp` and `devcheck` |
 | `api-mirror` | MirrorService: persistent self-refreshing local mirror (embedded SQLite + FTS5) of a bulk upstream dataset — Tier 3 opt-in |
@@ -346,11 +364,11 @@ When you complete a skill's checklist, check the boxes and add a completion time
 
 ## Bundling
 
-`npm run bundle` produces a `.mcpb` extension bundle for one-click install in Claude Desktop. MCPB is stdio-only — HTTP and Cloudflare Workers deployments are unaffected. Consumers who don't need it can delete `manifest.json` and `.mcpbignore`; `lint:packaging` skips cleanly.
+`npm run bundle` produces a `.mcpb` extension bundle for one-click install in Claude Desktop. The pack step is followed by `scripts/clean-mcpb.ts`, which prunes dev dependencies (`mcpb clean`) and strips two classes of `node_modules/**` content that root-anchored `.mcpbignore` patterns cannot reach: dependency-shipped agent docs (`framework-skills/`, `skills/`, `.claude/`, `.agents/`, `SKILL.md`) and platform-specific native bindings, which would otherwise lock the bundle to the platform it was packed on. MCPB is stdio-only — HTTP and Cloudflare Workers deployments are unaffected. Consumers who don't need it can delete `manifest.json` and `.mcpbignore`; `lint:packaging` skips cleanly.
 
 **Adding an env var requires both files:** `server.json` (registry discovery, `environmentVariables[]`) and `manifest.json` (bundle install UX, `mcp_config.env` + `user_config`). `lint:packaging` (run by `devcheck`) verifies the env var names match.
 
-**README install badges** (Claude Desktop `.mcpb`, Cursor, VS Code) and the `base64` / `encodeURIComponent` config-generation commands are ship-time concerns — run the `polish-docs-meta` skill, which carries the badge format, layout, and generation snippets in `skills/polish-docs-meta/references/readme.md`.
+**README install badges** (Claude Desktop `.mcpb`, Cursor, VS Code) and the `base64` / `encodeURIComponent` config-generation commands are ship-time concerns — run the `polish-docs-meta` skill, which carries the badge format, layout, and generation snippets in `framework-skills/polish-docs-meta/references/readme.md`.
 
 ---
 
@@ -364,23 +382,29 @@ Each per-version file opens with YAML frontmatter:
 ---
 summary: "One-line headline, ≤350 chars"  # required — powers the rollup index
 breaking: false                            # optional — true flags breaking changes
-security: false                            # optional — true flags security fixes
+security: false                            # optional — true ONLY for a source-code security fix, never a dependency CVE bump
 ---
 
 # 0.1.0 — YYYY-MM-DD
 ...
 ```
 
-`breaking: true` renders a `· ⚠️ Breaking` badge — use it when consumers must update code on upgrade (signature changes, removed APIs, config renames). `security: true` renders a `· 🛡️ Security` badge and pairs with a `## Security` body section. When both are set, badges render `· ⚠️ Breaking · 🛡️ Security`.
+`breaking: true` renders a `· ⚠️ Breaking` badge — use it when consumers must update code on upgrade (signature changes, removed APIs, config renames). `security: true` renders a `· 🛡️ Security` badge and pairs with a `## Security` body section — set it only for a security fix in this server's *own source code*, never for a routine dependency or transitive CVE bump (record those under `## Dependencies`). When both are set, badges render `· ⚠️ Breaking · 🛡️ Security`.
 
 `agent-notes` is an optional free-form field for maintenance agents processing the release downstream. Content here won't appear in the rendered CHANGELOG — it's consumed by agents running the `maintenance` skill. Use it for adoption instructions that don't fit the human-facing sections: new files to create, fields to populate, one-time migration steps. Omit entirely when there's nothing to say.
 
-**Section order** (Keep a Changelog): Added, Changed, Deprecated, Removed, Fixed, Security. Include only sections with entries — don't ship empty headers.
+**Section order:** the Keep a Changelog sequence — Added, Changed, Deprecated, Removed, Fixed, Security — then `Dependencies` last. Include only sections with entries — don't ship empty headers.
 
 **Tag annotations** render as GitHub Release bodies via `--notes-from-tag`. They must be structured markdown — never a flat comma-separated string. Subject omits the version number (GitHub prepends it). See `changelog/template.md` for the full format reference.
 
 ---
 
+## Publishing
+
+**Every release goes through a release PR, straight-through** — `git-wrapup`'s "Release PR mode", mode `straight-through`. One run: `git-wrapup` lands the commit stack on `release/<version>`, pushes it, and opens the PR (title = the release commit subject, body = the changelog entry plus a gates section); `release-and-publish` then fast-forwards `main` locally with `git merge --ff-only`, creates the tag on `main`'s tip, pushes `main` and the tag, deletes the branch, and publishes. A caller's brief may run a given release as `gated` instead — a `release-pr-review` pass on the open PR before `release-and-publish`. **Never merge through the GitHub UI or `gh pr merge`**: squash and rebase-merge are disabled in the repo settings because both rewrite the stack (rebase-merge also strips the SSH signatures), and a merge commit breaks the linear history.
+
+---
+
 ## Imports
 
 ```ts

package/Dockerfile

@@ -3,8 +3,14 @@
 #
 # This stage installs all dependencies (including dev), builds the TypeScript
 # source code into JavaScript, and prepares the production assets.
+#
+# Pinned to the builder's own platform. The build emits portable JavaScript, so
+# there is nothing to gain from running it once per target architecture — and
+# under a multi-arch build the emulated leg is actively broken: Bun 1.4.0 aborts
+# with SIGABRT inside qemu, so `bun run build` dies before it compiles anything.
+# The production stage below stays per-target and resolves its own dependencies.
 # ==============================================================================
-FROM oven/bun:1.3.14 AS build
+FROM --platform=$BUILDPLATFORM oven/bun:1.4.0 AS build
 
 WORKDIR /usr/src/app
 
@@ -30,7 +36,7 @@ RUN bun run build
 # application. It uses a slim base image and only includes production
 # dependencies and build artifacts.
 # ==============================================================================
-FROM oven/bun:1.3.14-slim AS production
+FROM oven/bun:1.4.0-slim AS production
 
 WORKDIR /usr/src/app
 
@@ -51,8 +57,13 @@ COPY package.json bun.lock ./
 
 # Install only production dependencies, ignoring any lifecycle scripts (like 'prepare')
 # that are not needed in the final production image.
+# `--omit=peer` drops the framework's optional peer tiers (test runner, service
+# SDKs, parsers) that Bun would otherwise auto-install. Anything this server
+# actually imports belongs in its own `dependencies`, so nothing needed at
+# runtime is lost. The OTEL step below carries the same flag — without it, that
+# install re-resolves the graph and pulls every optional peer back in.
 RUN --mount=type=cache,target=/root/.bun/install/cache \
-    bun install --production --frozen-lockfile --ignore-scripts
+    bun install --production --omit=peer --frozen-lockfile --ignore-scripts
 
 # Conditionally install OpenTelemetry optional peer dependencies (Tier 3).
 # These are not bundled by default to keep the base image lean. Enable at build time
@@ -60,7 +71,7 @@ RUN --mount=type=cache,target=/root/.bun/install/cache \
 ARG OTEL_ENABLED=true
 RUN --mount=type=cache,target=/root/.bun/install/cache \
     if [ "$OTEL_ENABLED" = "true" ]; then \
-      bun add --omit=dev --ignore-scripts @hono/otel \
+      bun add --omit=dev --omit=peer --ignore-scripts @hono/otel \
         @opentelemetry/instrumentation-http \
         @opentelemetry/exporter-metrics-otlp-http \
         @opentelemetry/exporter-trace-otlp-http \

package/README.md

@@ -7,7 +7,7 @@
 
 <div align="center">
 
-[![Version](https://img.shields.io/badge/Version-0.2.0-blue.svg?style=flat-square)](./CHANGELOG.md) [![License](https://img.shields.io/badge/License-Apache%202.0-orange.svg?style=flat-square)](./LICENSE) [![Docker](https://img.shields.io/badge/Docker-ghcr.io-2496ED?style=flat-square&logo=docker&logoColor=white)](https://github.com/users/cyanheads/packages/container/package/nist-nvd-mcp-server) [![MCP SDK](https://img.shields.io/badge/MCP%20SDK-^1.29.0-green.svg?style=flat-square)](https://modelcontextprotocol.io/) [![npm](https://img.shields.io/npm/v/@cyanheads/nist-nvd-mcp-server?style=flat-square&logo=npm&logoColor=white)](https://www.npmjs.com/package/@cyanheads/nist-nvd-mcp-server) [![TypeScript](https://img.shields.io/badge/TypeScript-^6.0.3-3178C6.svg?style=flat-square)](https://www.typescriptlang.org/) [![Bun](https://img.shields.io/badge/Bun-v1.3.14-blueviolet.svg?style=flat-square)](https://bun.sh/)
+[![Version](https://img.shields.io/badge/Version-0.3.1-blue.svg?style=flat-square)](./CHANGELOG.md) [![License](https://img.shields.io/badge/License-Apache%202.0-orange.svg?style=flat-square)](./LICENSE) [![Docker](https://img.shields.io/badge/Docker-ghcr.io-2496ED?style=flat-square&logo=docker&logoColor=white)](https://github.com/users/cyanheads/packages/container/package/nist-nvd-mcp-server) [![MCP SDK](https://img.shields.io/badge/MCP%20SDK-^2.0.0-green.svg?style=flat-square)](https://modelcontextprotocol.io/) [![npm](https://img.shields.io/npm/v/@cyanheads/nist-nvd-mcp-server?style=flat-square&logo=npm&logoColor=white)](https://www.npmjs.com/package/@cyanheads/nist-nvd-mcp-server) [![TypeScript](https://img.shields.io/badge/TypeScript-^7.0.2-3178C6.svg?style=flat-square)](https://www.typescriptlang.org/) [![Bun](https://img.shields.io/badge/Bun-v1.4.0-blueviolet.svg?style=flat-square)](https://bun.sh/)
 
 </div>
 
@@ -27,9 +27,11 @@
 
 ---
 
-## Tools
+## Overview
 
-Five tools for vulnerability research, CPE auditing, and change tracking against the NIST NVD API 2.0:
+CVE and CPE data from the NIST National Vulnerability Database. Search and audit vulnerabilities by keyword, severity, CWE, or CISA KEV status, resolve products to CPE names, and track a CVE's revision history from any MCP client. Runs as a stdio process, a local Streamable HTTP server, or the public hosted endpoint above.
+
+### Tools
 
 | Tool | Description |
 |:-----|:------------|
@@ -39,106 +41,108 @@ Five tools for vulnerability research, CPE auditing, and change tracking against
 | `nvd_audit_cpe` | Find all CVEs affecting a specific product version by CPE name or virtual match string. |
 | `nvd_get_cve_history` | Retrieve the change history for a CVE — score revisions, status transitions, and reference additions. |
 
-### `nvd_search_cves`
+### Resources
 
-The primary discovery tool for vulnerability surveillance and triage workflows.
+| Resource | Description |
+|:---------|:------------|
+| `nvd://cve/{cveId}` | Full CVE record by ID — same data as `nvd_get_cve` for a single ID, as a stable URI for injectable context. |
 
-- Full-text keyword search across CVE descriptions (AND-semantics across words), or `exactPhrase: true` to match the keyword as a phrase
-- Severity filter by CVSS v2/v3/v4 label (LOW, MEDIUM, HIGH, CRITICAL)
-- CWE weakness filter (e.g., `CWE-79`, `NVD-CWE-Other`)
-- CISA KEV filter — limit results to known-exploited vulnerabilities
-- Convenience date shorthands: `pubDays` and `lastModDays` for "last N days" queries
-- Explicit ISO 8601 date range parameters (`pubStartDate`/`pubEndDate`, etc.) with 120-day max span
-- Auto-clamps convenience date params that exceed 120 days and reports clamped values in the response enrichment
-- Pagination via `limit` (up to 2000) and `offset`
-- Every row carries a truncated description alongside the ID, so results are distinguishable without a follow-up fetch
-- Results are always brief; call `nvd_get_cve` for full detail
+All resource data is also reachable via tools.
 
----
+## Capability reference
 
-### `nvd_get_cve`
+### `nvd_search_cves` <sub>tool</sub>
+
+- Full-text keyword search (AND-semantics across words), or `exactPhrase: true` for an exact-phrase match — requires `keyword`
+- Filters: CVSS severity band (LOW/MEDIUM/HIGH/CRITICAL — CRITICAL requires `severityVersion: "v3"` or `"v4"`), CWE ID, CISA KEV status, `noRejected` (default true)
+- Date filters: `pubDays`/`lastModDays` convenience shorthands (auto-clamped to 120 days, clamping reported in the enrichment) or explicit ISO 8601 ranges (120-day max span, both ends required); the two forms per axis are mutually exclusive
+- Pagination via `limit` (up to 2000, default 20) and `offset`
+- Always returns brief summaries with a truncated description; call `nvd_get_cve` for full detail
+
+---
 
-Fetch one or more CVEs by ID with full detail or brief summaries.
+### `nvd_get_cve` <sub>tool</sub>
 
 - Batch up to 100 CVE IDs per call
-- Full mode: all CVSS scores across v2.0, v3.0, v3.1, and v4.0; CWE weaknesses; CPE configurations; CISA KEV fields; references
-- Brief mode (`brief: true`): ID, status, top severity, KEV name, truncated description — recommended for batches larger than 10
-- `includeReferences: false` to strip the references array and reduce response size
-- Per-ID parity check: the `missingIds` enrichment field lists any requested IDs NVD didn't return
-- Rendered text carries the affected-product criteria and references the record holds, capped with a `… N more` trailer; `allLanguages: true` renders every localized description, not just English
+- Full mode (default): CVSS scores across v2.0/v3.0/v3.1/v4.0, CWE weaknesses, CPE configurations, CISA KEV fields, references
+- `brief: true` returns trimmed rows (ID, status, top severity, KEV name, truncated description) — recommended for batches over 10
+- `includeReferences: false` strips the references array; `allLanguages: true` renders every localized description instead of English-only
+- `missingIds` enrichment field lists any requested IDs NVD didn't return
+- Rendered text caps references at 15 per record, with a `… N more` trailer
 
 ---
 
-### `nvd_search_cpes`
+### `nvd_search_cpes` <sub>tool</sub>
 
-Look up product identifiers before auditing.
-
-- Keyword search (e.g., `"apache http server"`, `"openssl"`) or partial CPEv2.3 pattern
+- Keyword search (e.g. `"apache http server"`) or a partial CPEv2.3 pattern via `cpeMatchString` — at least one required
 - Returns full CPE name, human-readable title, deprecation status, and superseding CPEs
-- Pagination via `limit` (up to 10,000 per page) and `offset` — a vendor-level keyword can match tens of thousands of entries, so page with `offset` rather than trying to narrow further
-- Use this before `nvd_audit_cpe` — CPE names are arcane strings; guessing audits the wrong product
+- Pagination via `limit` (up to 10,000, default 20) and `offset` — a vendor-level keyword can match tens of thousands of entries, so page rather than narrowing further
+- Use before `nvd_audit_cpe` to resolve the exact CPE name a product needs
 
 ---
 
-### `nvd_audit_cpe`
-
-Full CVE audit for a specific product version.
+### `nvd_audit_cpe` <sub>tool</sub>
 
-- Two modes: exact `cpeName` (NVD auto-applies `isVulnerable`) or `virtualMatchString` with optional version range bounds
-- Version range via `versionStart`/`versionEnd` with inclusive/exclusive type control
-- Client-side severity filter (`severityMin`) to strip low-signal entries
-- Returns full CVE records (ID, CVSS scores, CWE, CPE configurations, KEV fields, references)
-- Pagination via `limit` (up to 2000) and `offset` — page at a modest `limit` instead of raising it, since each result is a full record
-- Echoes the CPE identifier used in the response enrichment so callers can verify the correct product was queried
+- Two modes: exact `cpeName` (NVD auto-applies `isVulnerable`) or `virtualMatchString` with optional `versionStart`/`versionEnd` bounds (inclusive/exclusive)
+- Client-side `severityMin` filter drops low-signal entries from the fetched page — it can only remove what `limit` already retrieved
+- Returns full CVE records (CVSS scores, CWE, CPE configurations, KEV fields, references)
+- Pagination via `limit` (up to 2000, default 20) and `offset` — page at a modest limit rather than raising it, since each result is a full record
+- `auditTarget` enrichment field echoes the CPE identifier used, so callers can verify the correct product was queried
 
 ---
 
-### `nvd_get_cve_history`
-
-Track a CVE's lifecycle over time.
+### `nvd_get_cve_history` <sub>tool</sub>
 
 - Returns change events: CVSS revisions, status transitions, reference additions, CPE configuration updates
-- `order` picks which end to read from — `newest` (default) returns the most recent events first, `oldest` returns NVD's native oldest-first order
-- Paginated via `limit` and `offset`, where `offset` counts from the end `order` anchors to
-- Note: the NVD history endpoint is significantly slower without an API key — set `NVD_API_KEY` and raise `NVD_REQUEST_TIMEOUT_MS` for reliable operation
+- `order` picks the anchor end — `newest` (default) reads most-recent-first, `oldest` reads NVD's native order
+- Paginated via `limit` (up to 2000, default 20) and `offset`, counted from the end `order` anchors to
+- The history endpoint is markedly slower without an API key — set `NVD_API_KEY` and raise `NVD_REQUEST_TIMEOUT_MS`
 
-## Resource
+---
 
-| Type | Name | Description |
-|:-----|:-----|:------------|
-| Resource | `nvd://cve/{cveId}` | Full CVE record by ID — same data as `nvd_get_cve` for a single ID, as a stable URI for injectable context. |
+### `nvd://cve/{cveId}` <sub>resource</sub>
 
-All resource data is also reachable via tools.
+- Full CVE record as `application/json` — same data as `nvd_get_cve` for one ID, with references and English-only descriptions
+- `cveId` must match `CVE-YYYY-NNNNN`; a well-formed but unknown ID throws `cve_not_found`
 
 ## Features
 
-Built on [`@cyanheads/mcp-ts-core`](https://www.npmjs.com/package/@cyanheads/mcp-ts-core):
-
-- Declarative tool, resource, and prompt definitions — single file per primitive, framework handles registration and validation
-- Unified error handling — handlers throw, framework catches, classifies, and formats
-- Pluggable auth: `none`, `jwt`, `oauth`
-- Swappable storage backends: `in-memory`, `filesystem`, `Supabase`, `Cloudflare KV/R2/D1`
-- Structured logging with optional OpenTelemetry tracing
-- STDIO and Streamable HTTP transports
+Built on [`@cyanheads/mcp-ts-core`](https://github.com/cyanheads/mcp-ts-core): stdio and Streamable HTTP transports, pluggable auth (`none` / `jwt` / `oauth`), swappable storage (`in-memory`, `filesystem`, `Supabase`, `Cloudflare KV/R2/D1`), structured logging with optional OpenTelemetry tracing.
 
 NVD-specific:
 
 - Request pacer enforces NVD's 5 req/30s (no key) and 50 req/30s (with key) limits with automatic queuing, at a minimum inter-request gap derived from the window and limit
 - Retry wraps the pacer rather than sitting inside it — every attempt takes its own turn in the queue, so retries count against the rate budget instead of bursting past it
-- A 403's `Retry-After` holds the whole queue until NVD's window resets. Keyless, a 403 fails fast and names `NVD_API_KEY` rather than spending a 5-request budget on retries that cannot outlast a 30-second window
-- Deterministic rejections fail fast instead of consuming retries. NVD answers both a bad parameter and a refused API key with HTTP 404, separated only by a `message` header — a refused key surfaces as a config fault naming `NVD_API_KEY` rather than as a malformed CVE ID
-- HTML-response guard catches NVD rate-limit pages served as HTML instead of 403
+- A 403's `Retry-After` holds the whole queue until NVD's window resets; keyless, a 403 fails fast and names `NVD_API_KEY` rather than spending the 5-request budget on retries that cannot outlast a 30-second window
+- Deterministic rejections fail fast instead of consuming retries — NVD answers both a bad parameter and a refused API key with HTTP 404, separated only by a `message` header, so a refused key surfaces as a config fault naming `NVD_API_KEY` rather than as a malformed CVE ID
+- HTML-response guard catches NVD rate-limit pages served as HTML instead of a 403
 
 Agent-friendly output:
 
 - An `enrichment` block on every response, carried on both `structuredContent` and the rendered text — total results, returned count, page offset, the filters actually applied, and any date-clamping events, so agents can reason about what was really queried
-- `missingIds` in batch CVE lookups — per-ID parity check instead of a silent partial result
+- `missingIds` in batch CVE lookups — a per-ID parity check instead of a silent partial result
 - CPE echo in audit responses — `cpeName` or `virtualMatchString` reflected back so callers can verify the correct product was audited
-- Empty-result notices that name the cause — an unmatched query, a severity threshold that emptied the page, and an offset past the end of the result set are told apart rather than all reading as "nothing found"
-- An audit that finds nothing is a result, not an error — a product with no CVEs in NVD returns an empty page with `totalCount: 0` on either input arm, so "no known vulnerabilities" reads as the answer it is
+- Empty-result notices that name the cause — an unmatched query, a severity threshold that emptied the page, an offset past the end of the result set, and a clean audit ("no known vulnerabilities") are told apart rather than all reading as errors or "nothing found"
 
 ## Getting started
 
+### Public Hosted Instance
+
+A public instance is available at `https://nist-nvd.caseyjhand.com/mcp` — no installation required. Point any MCP client at it via Streamable HTTP:
+
+```json
+{
+  "mcpServers": {
+    "nist-nvd-mcp-server": {
+      "type": "streamable-http",
+      "url": "https://nist-nvd.caseyjhand.com/mcp"
+    }
+  }
+}
+```
+
+### Self-Hosted / Local
+
 Add the following to your MCP client configuration file.
 
 ```json
@@ -205,7 +209,7 @@ MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 NVD_API_KEY=... bun run start:http
 
 ### Prerequisites
 
-- [Bun v1.3.0](https://bun.sh/) or higher (or Node.js v24+).
+- [Bun v1.4.0](https://bun.sh/) or higher (or Node.js v24+).
 - Optional: [NVD API key](https://nvd.nist.gov/developers/request-an-api-key) — free, raises rate limit from 5 req/30s to 50 req/30s.
 
 ### Installation
@@ -308,7 +312,7 @@ See [`CLAUDE.md`](./CLAUDE.md) for development guidelines and architectural rule
 
 ## Contributing
 
-Issues and pull requests are welcome. Run checks and tests before submitting:
+Issues are welcome. Run checks and tests before submitting:
 
 ```sh
 bun run devcheck