Strengthen release verification gates

2026-06-25 07:05:08 -07:00
parent 3d7dca8773
commit 99881b608b
7 changed files with 298 additions and 20 deletions
--- a/bin/context-kit
+++ b/bin/context-kit
@@ -42,9 +42,10 @@ fail() {

 load_env_file

+DEFAULT_DATA_DIR="${HOME:-${PWD}}/.local/share/context-kit"
 PROJECT="${CONTEXT_KIT_COMPOSE_PROJECT:-context-kit}"
 COMPOSE_FILE="${ROOT}/compose.yml"
-DATA_DIR="${CONTEXT_KIT_DATA_DIR:-${HOME}/.local/share/context-kit}"
+DATA_DIR="${CONTEXT_KIT_DATA_DIR:-${DEFAULT_DATA_DIR}}"
 NETWORK="${CONTEXT_KIT_DOCKER_NETWORK:-${PROJECT}_default}"
 SEARXNG_PORT="${CONTEXT_KIT_SEARXNG_PORT:-8099}"
 DOCS_PORT="${CONTEXT_KIT_DOCS_PORT:-8776}"
@@ -190,6 +191,17 @@ warn() {
  printf 'warn: %s\n' "$*" >&2
 }

+print_relative_paths() {
+  local path
+  while IFS= read -r path; do
+    [[ -n "${path}" ]] || continue
+    if [[ "${path}" == "${ROOT}/"* ]]; then
+      path="${path#"${ROOT}/"}"
+    fi
+    printf '%s\n' "${path}"
+  done
+}
+
 json_escape() {
  local s="$1"
  s="${s//\\/\\\\}"
@@ -314,8 +326,9 @@ cmd_status() {
  printf '\nImages\n'
  docker image ls --format '{{.Repository}}:{{.Tag}}\t{{.Size}}' \
    | grep -E '^(context-kit/|ghcr.io/yamadashy/repomix:)' || true
-  printf '\nLabeled containers\n'
-  docker ps -a --filter label=dev.context-kit=true --format 'table {{.Names}}\t{{.Status}}\t{{.Image}}'
+  printf '\nActive per-call MCP containers\n'
+  docker ps -a --filter label=dev.context-kit=true --format '{{.Names}}\t{{.Status}}\t{{.Image}}\t{{.Command}}' \
+    | awk 'BEGIN { print "NAMES\tSTATUS\tIMAGE\tCOMMAND" } $1 !~ /^context-kit-(docs-mcp|searxng-1)$/ { print }'
  printf '\nDocs MCP endpoint\n- %s (container: %s)\n' "${DOCS_HTTP_URL}" "${DOCS_CONTAINER_NAME}"
  printf '\nDocs sources\n'
  resolved_sources | sed 's/^/- /'
@@ -387,13 +400,15 @@ cmd_doctor() {
  if command -v curl >/dev/null 2>&1 && curl -fsS "http://127.0.0.1:${SEARXNG_PORT}/healthz" >/dev/null 2>&1; then
    printf 'pass SearXNG responds on 127.0.0.1:%s\n' "${SEARXNG_PORT}"
  else
-    printf 'warn SearXNG not responding on 127.0.0.1:%s\n' "${SEARXNG_PORT}"
+    printf 'fail SearXNG not responding on 127.0.0.1:%s\n' "${SEARXNG_PORT}"
+    ok=1
  fi

  if command -v curl >/dev/null 2>&1 && curl -fsS -o /dev/null "http://127.0.0.1:${DOCS_PORT}/status" 2>/dev/null; then
    printf 'pass docs-mcp HTTP responds on 127.0.0.1:%s\n' "${DOCS_PORT}"
  else
-    printf 'warn docs-mcp HTTP not responding on 127.0.0.1:%s (run context-kit start)\n' "${DOCS_PORT}"
+    printf 'fail docs-mcp HTTP not responding on 127.0.0.1:%s (run context-kit start)\n' "${DOCS_PORT}"
+    ok=1
  fi

  if [[ "$(resolved_sources | wc -l | tr -d ' ')" -gt 0 ]]; then
@@ -409,8 +424,13 @@ cmd_web_search() {
  require_docker
  require_network
  require_image "${WEB_SEARCH_IMAGE}" "context-kit build"
+  local cidfile_args=()
+  if [[ -n "${CONTEXT_KIT_DOCKER_CIDFILE:-}" ]]; then
+    cidfile_args=(--cidfile "${CONTEXT_KIT_DOCKER_CIDFILE}")
+  fi
  exec docker run --rm -i \
    --label dev.context-kit=true \
+    "${cidfile_args[@]}" \
    --network "${NETWORK}" \
    -e DEFAULT_SEARCH_PROVIDER="${WEB_SEARCH_PROVIDER}" \
    -e SEARXNG_URL="${SEARXNG_URL:-http://searxng:8080}" \
@@ -438,8 +458,13 @@ cmd_docs() {
  fi

  local bridge_url="http://${DOCS_CONTAINER_NAME}:8000/mcp"
+  local cidfile_args=()
+  if [[ -n "${CONTEXT_KIT_DOCKER_CIDFILE:-}" ]]; then
+    cidfile_args=(--cidfile "${CONTEXT_KIT_DOCKER_CIDFILE}")
+  fi
  exec docker run --rm -i \
    --label dev.context-kit=true \
+    "${cidfile_args[@]}" \
    --network "${NETWORK}" \
    --entrypoint mcp-proxy \
    "${DOCS_IMAGE}" \
@@ -454,8 +479,13 @@ cmd_repomix() {
  dir="$(project_dir)"
  mount_dir="${CONTEXT_KIT_REPOMIX_MOUNT_DIR:-${dir}}"
  mount_dir="$(cd "${mount_dir}" && pwd -P)"
+  local cidfile_args=()
+  if [[ -n "${CONTEXT_KIT_DOCKER_CIDFILE:-}" ]]; then
+    cidfile_args=(--cidfile "${CONTEXT_KIT_DOCKER_CIDFILE}")
+  fi
  exec docker run --rm -i \
    --label dev.context-kit=true \
+    "${cidfile_args[@]}" \
    -v "${mount_dir}:${mount_dir}:ro" \
    --workdir "${dir}" \
    "${REPOMIX_IMAGE}" --mcp
@@ -536,6 +566,10 @@ cmd_install() {

 cmd_redaction_check() {
  local bad=0
+  local scan_paths=("${ROOT}")
+  if [[ "$#" -gt 0 ]]; then
+    scan_paths=("$@")
+  fi
  local local_path_terms='/(home|Users)/[^/[:space:]]+|/data/(projects|opencode-mcp)[^[:space:]]*|[A-Za-z]:\\Users\\[^\\[:space:]]+'
  local secret_terms='AKIA[0-9A-Z]{16}|BEGIN (RSA |OPENSSH |EC |DSA )?PRIVATE KEY|xox[baprs]-|sk-[A-Za-z0-9_-]{20,}|ghp_[A-Za-z0-9_]{20,}|github_pat_[A-Za-z0-9_]{20,}|glpat-[A-Za-z0-9_-]{20,}|gitea_[A-Za-z0-9_-]{20,}'

@@ -551,10 +585,18 @@ cmd_redaction_check() {
    --exclude=*.log
  )

-  if grep "${grep_opts[@]}" "${local_path_terms}" "${ROOT}"; then
+  local matches
+  matches="$(grep "${grep_opts[@]}" --files-with-matches "${local_path_terms}" "${scan_paths[@]}" 2>/dev/null || true)"
+  if [[ -n "${matches}" ]]; then
+    printf 'fail redaction-check found local path patterns in:\n' >&2
+    printf '%s\n' "${matches}" | print_relative_paths | sed 's/^/- /' >&2
    bad=1
  fi
-  if grep "${grep_opts[@]}" "${secret_terms}" "${ROOT}"; then
+
+  matches="$(grep "${grep_opts[@]}" --files-with-matches "${secret_terms}" "${scan_paths[@]}" 2>/dev/null || true)"
+  if [[ -n "${matches}" ]]; then
+    printf 'fail redaction-check found secret-like patterns in:\n' >&2
+    printf '%s\n' "${matches}" | print_relative_paths | sed 's/^/- /' >&2
    bad=1
  fi

--- a/compose.yml
+++ b/compose.yml
@@ -62,10 +62,10 @@ services:
      # refresh on demand. Set CONTEXT_KIT_DOCS_PREINDEX=1 to restore eager.
      DOCS_MCP_PREINDEX: "${CONTEXT_KIT_DOCS_PREINDEX:-0}"
    volumes:
-      - ${CONTEXT_KIT_DATA_DIR:-${HOME}/.local/share/context-kit}/docs:/data
-      - ${CONTEXT_KIT_DATA_DIR:-${HOME}/.local/share/context-kit}/models:/models
-      - ${CONTEXT_KIT_DATA_DIR:-${HOME}/.local/share/context-kit}/docs-sources.txt:/etc/context-kit/docs-sources.txt:ro
-      - ${CONTEXT_KIT_DOCS_LOCAL_SOURCES_DIR:-${CONTEXT_KIT_DATA_DIR:-${HOME}/.local/share/context-kit}/local-sources}:/etc/context-kit/local-sources:ro
+      - ${CONTEXT_KIT_DATA_DIR:-${HOME:?Set HOME or CONTEXT_KIT_DATA_DIR, or run bin/context-kit instead}/.local/share/context-kit}/docs:/data
+      - ${CONTEXT_KIT_DATA_DIR:-${HOME:?Set HOME or CONTEXT_KIT_DATA_DIR, or run bin/context-kit instead}/.local/share/context-kit}/models:/models
+      - ${CONTEXT_KIT_DATA_DIR:-${HOME:?Set HOME or CONTEXT_KIT_DATA_DIR, or run bin/context-kit instead}/.local/share/context-kit}/docs-sources.txt:/etc/context-kit/docs-sources.txt:ro
+      - ${CONTEXT_KIT_DOCS_LOCAL_SOURCES_DIR:-${CONTEXT_KIT_DATA_DIR:-${HOME:?Set HOME or CONTEXT_KIT_DATA_DIR, or run bin/context-kit instead}/.local/share/context-kit}/local-sources}:/etc/context-kit/local-sources:ro
    healthcheck:
      test: ["CMD-SHELL", "python -c \"import urllib.request,sys; sys.exit(0 if urllib.request.urlopen('http://127.0.0.1:8000/status', timeout=2).status < 500 else 1)\""]
      interval: 30s
--- a/docker/docs/entrypoint.sh
+++ b/docker/docs/entrypoint.sh
@@ -21,14 +21,24 @@ if [ ! -r "$sources_file" ]; then
  exit 64
 fi

-# Strip comments and blank lines, then collapse whitespace into a flat list.
-sources=$(grep -vE '^[[:space:]]*(#|$)' "$sources_file" | tr -s '[:space:]' '\n' | grep -v '^$' || true)
+# Strip inline comments and blank lines, then collapse whitespace into a flat list.
+sources=$(sed 's/#.*//' "$sources_file" | tr -s '[:space:]' '\n' | grep -v '^$' || true)

 if [ -z "$sources" ]; then
  echo "docs-mcp: no sources found in $sources_file after stripping comments/blanks" >&2
  exit 64
 fi

+for source_url in $sources; do
+  case "$source_url" in
+    */llms.txt|*/llms-full.txt) ;;
+    *)
+      echo "docs-mcp: source URL must end with /llms.txt or /llms-full.txt: $source_url" >&2
+      exit 64
+      ;;
+  esac
+done
+
 if [ -d "$local_sources_dir" ]; then
  python -m http.server "$local_sources_port" \
    --bind 127.0.0.1 \
--- a/docs/configuration.md
+++ b/docs/configuration.md
@@ -78,5 +78,6 @@ Each source file is plain text. Blank lines and `#` comments are ignored.
 Entries may be absolute source-profile paths for private machine-local config.
 For local llms.txt files, place content under
 `CONTEXT_KIT_DOCS_LOCAL_SOURCES_DIR` and reference it as
-`http://127.0.0.1:8769/path/inside/local-sources.txt`; that loopback URL is
-inside the docs-mcp container, not exposed on the host.
+`http://127.0.0.1:8769/path/inside/local-sources/llms.txt` or another URL that
+ends in `/llms.txt` or `/llms-full.txt`; that loopback URL is inside the docs-mcp
+container, not exposed on the host.
--- a/scripts/release-check
+++ b/scripts/release-check
@@ -17,17 +17,34 @@ check_node() {
  done
 }

-git diff --check
+assert_redaction_check_does_not_disclose_matches() {
+  local fixture="${tmp_dir}/redaction-fixture.txt"
+  local output="${tmp_dir}/redaction-output.txt"
+  local blocked_path="/data/proj""ects/context-kit-private-fixture"
+  printf 'blocked=%s\n' "${blocked_path}" > "${fixture}"
+  if bin/context-kit redaction-check "${fixture}" >"${output}" 2>&1; then
+    printf 'redaction-check test unexpectedly passed\n' >&2
+    return 1
+  fi
+  if grep -F "${blocked_path}" "${output}" >/dev/null; then
+    printf 'redaction-check disclosed matched content\n' >&2
+    return 1
+  fi
+}
+
+git diff --check HEAD
+git show --check --format= HEAD >/dev/null
 git ls-files --cached --error-unmatch \
  docker/web-search/patch-mcp-web-search.mjs \
  docker/web-search/overrides/bing.js \
  docker/docs/constraints.txt \
  scripts/smoke-web-search.mjs \
+  scripts/smoke-docs.mjs \
  scripts/release-check >/dev/null
 bash -n bin/context-kit
 bash -n scripts/release-check
 sh -n docker/docs/entrypoint.sh
-check_node docker/web-search/patch-mcp-web-search.mjs docker/web-search/overrides/bing.js scripts/smoke-web-search.mjs
+check_node docker/web-search/patch-mcp-web-search.mjs docker/web-search/overrides/bing.js scripts/smoke-web-search.mjs scripts/smoke-docs.mjs

 node -e 'const fs=require("node:fs"); JSON.parse(fs.readFileSync("snippets/opencode.json", "utf8")); JSON.parse(fs.readFileSync("snippets/claude.mcp.json", "utf8"));'
 bin/context-kit install opencode > "${tmp_dir}/opencode.json"
@@ -39,11 +56,20 @@ node -e 'const fs=require("node:fs"); for (const file of process.argv.slice(1))
  "${tmp_dir}/opencode-absolute.json" \
  "${tmp_dir}/claude.json" \
  "${tmp_dir}/claude-absolute.json"
+bin/context-kit redaction-check "${tmp_dir}/opencode.json" "${tmp_dir}/claude.json"
+assert_redaction_check_does_not_disclose_matches

 bin/context-kit redaction-check
 docker compose -p context-kit -f compose.yml config >/dev/null
+if env -u HOME docker compose --env-file /dev/null -p context-kit-release-home-check -f compose.yml config >"${tmp_dir}/compose-no-home.out" 2>"${tmp_dir}/compose-no-home.err"; then
+  printf 'compose config unexpectedly succeeded without HOME or CONTEXT_KIT_DATA_DIR\n' >&2
+  exit 1
+fi
+CONTEXT_KIT_DATA_DIR="${tmp_dir}/compose-data" env -u HOME docker compose --env-file /dev/null -p context-kit-release-home-check -f compose.yml config >/dev/null
 bin/context-kit build
+bin/context-kit restart
 bin/context-kit doctor
 node scripts/smoke-web-search.mjs bin/context-kit web-search
+node scripts/smoke-docs.mjs bin/context-kit docs

 printf 'pass release-check\n'
--- a/scripts/smoke-docs.mjs
+++ b/scripts/smoke-docs.mjs
@@ -0,0 +1,170 @@
+import { spawn, spawnSync } from "node:child_process";
+import { existsSync, mkdtempSync, readFileSync, rmSync } from "node:fs";
+import { tmpdir } from "node:os";
+import { join } from "node:path";
+
+const command = process.argv[2];
+const args = process.argv.slice(3);
+
+if (!command) {
+  throw new Error("usage: node scripts/smoke-docs.mjs <command> [args...]");
+}
+
+const tmpDir = mkdtempSync(join(tmpdir(), "context-kit-docs-smoke-"));
+const cidFile = join(tmpDir, "container.cid");
+
+const child = spawn(command, args, {
+  cwd: new URL("..", import.meta.url).pathname,
+  env: { ...process.env, CONTEXT_KIT_DOCKER_CIDFILE: cidFile },
+  stdio: ["pipe", "pipe", "pipe"]
+});
+
+let nextId = 1;
+const pending = new Map();
+let stdoutBuffer = "";
+let stderrBuffer = "";
+let childExited = false;
+
+child.once("exit", (code, signal) => {
+  childExited = true;
+  if (pending.size > 0) {
+    const error = new Error(`MCP child exited before responding (code=${code}, signal=${signal}). stderr: ${stderrBuffer.slice(-2000)}`);
+    for (const { reject } of pending.values()) reject(error);
+    pending.clear();
+  }
+});
+
+function stopChild() {
+  return new Promise(resolve => {
+    if (childExited) {
+      stopContainer();
+      rmSync(tmpDir, { recursive: true, force: true });
+      resolve();
+      return;
+    }
+
+    child.stdin.end();
+    const stopTimer = setTimeout(() => {
+      stopContainer();
+    }, 1000);
+    const termTimer = setTimeout(() => {
+      if (!childExited) child.kill("SIGTERM");
+    }, 3000);
+    const killTimer = setTimeout(() => {
+      if (!childExited) child.kill("SIGKILL");
+    }, 6000);
+
+    child.once("exit", () => {
+      stopContainer();
+      clearTimeout(stopTimer);
+      clearTimeout(termTimer);
+      clearTimeout(killTimer);
+      rmSync(tmpDir, { recursive: true, force: true });
+      resolve();
+    });
+  });
+}
+
+function stopContainer() {
+  if (!existsSync(cidFile)) return;
+  const containerId = readFileSync(cidFile, "utf8").trim();
+  if (!containerId) return;
+  spawnSync("docker", ["stop", containerId], { stdio: "ignore" });
+}
+
+const timeout = setTimeout(async () => {
+  await stopChild();
+  console.error(`Docs MCP smoke timed out. stderr: ${stderrBuffer.slice(-2000)}`);
+  process.exit(1);
+}, 300000);
+
+child.stderr.on("data", chunk => {
+  stderrBuffer += chunk.toString();
+});
+
+child.stdout.on("data", chunk => {
+  stdoutBuffer += chunk.toString();
+  let newline;
+  while ((newline = stdoutBuffer.indexOf("\n")) >= 0) {
+    const line = stdoutBuffer.slice(0, newline).trim();
+    stdoutBuffer = stdoutBuffer.slice(newline + 1);
+    if (!line) continue;
+    let message;
+    try {
+      message = JSON.parse(line);
+    } catch {
+      continue;
+    }
+    if (message.id && pending.has(message.id)) {
+      const { resolve, reject } = pending.get(message.id);
+      pending.delete(message.id);
+      if (message.error) reject(new Error(JSON.stringify(message.error)));
+      else resolve(message.result);
+    }
+  }
+});
+
+function request(method, params = {}) {
+  if (childExited) {
+    return Promise.reject(new Error(`MCP child already exited. stderr: ${stderrBuffer.slice(-2000)}`));
+  }
+  const id = nextId++;
+  child.stdin.write(`${JSON.stringify({ jsonrpc: "2.0", id, method, params })}\n`);
+  return new Promise((resolve, reject) => pending.set(id, { resolve, reject }));
+}
+
+function notify(method, params = {}) {
+  child.stdin.write(`${JSON.stringify({ jsonrpc: "2.0", method, params })}\n`);
+}
+
+async function callTool(name, args = {}) {
+  return request("tools/call", { name, arguments: args });
+}
+
+try {
+  await request("initialize", {
+    protocolVersion: "2024-11-05",
+    capabilities: {},
+    clientInfo: { name: "context-kit-docs-smoke", version: "0.0.0" }
+  });
+  notify("notifications/initialized");
+
+  const listed = await request("tools/list");
+  const toolNames = new Set((listed.tools || []).map(tool => tool.name));
+  for (const name of ["docs_query", "docs_sources"]) {
+    if (!toolNames.has(name)) throw new Error(`missing tool: ${name}`);
+  }
+
+  const sources = await callTool("docs_sources");
+  const sourcesText = JSON.stringify(sources);
+  if (sources.isError) {
+    throw new Error(`docs_sources returned an error: ${sourcesText.slice(0, 500)}`);
+  }
+
+  const query = await callTool("docs_query", {
+    query: "Model Context Protocol documentation",
+    limit: 3,
+    auto_retrieve: true,
+    auto_retrieve_threshold: 0.1,
+    auto_retrieve_limit: 1,
+    max_bytes: 12000
+  });
+  const queryText = JSON.stringify(query);
+  if (!queryText.includes("search_results") && !queryText.includes("Model Context Protocol")) {
+    throw new Error(`docs_query returned unexpected payload: ${queryText.slice(0, 500)}`);
+  }
+
+  clearTimeout(timeout);
+  await stopChild();
+  console.log(JSON.stringify({
+    tools: Array.from(toolNames).sort(),
+    docs_sources: "pass",
+    docs_query: "pass"
+  }, null, 2));
+} catch (error) {
+  clearTimeout(timeout);
+  await stopChild();
+  console.error(error.message);
+  if (stderrBuffer) console.error(stderrBuffer.slice(-4000));
+  process.exit(1);
+}
--- a/scripts/smoke-web-search.mjs
+++ b/scripts/smoke-web-search.mjs
@@ -1,4 +1,7 @@
-import { spawn } from "node:child_process";
+import { spawn, spawnSync } from "node:child_process";
+import { existsSync, mkdtempSync, readFileSync, rmSync } from "node:fs";
+import { tmpdir } from "node:os";
+import { join } from "node:path";

 const command = process.argv[2];
 const args = process.argv.slice(3);
@@ -7,9 +10,12 @@ if (!command) {
  throw new Error("usage: node scripts/smoke-web-search.mjs <command> [args...]");
 }

+const tmpDir = mkdtempSync(join(tmpdir(), "context-kit-web-search-smoke-"));
+const cidFile = join(tmpDir, "container.cid");
+
 const child = spawn(command, args, {
  cwd: new URL("..", import.meta.url).pathname,
-  env: process.env,
+  env: { ...process.env, CONTEXT_KIT_DOCKER_CIDFILE: cidFile },
  stdio: ["pipe", "pipe", "pipe"]
 });

@@ -19,18 +25,28 @@ let stdoutBuffer = "";
 let stderrBuffer = "";
 let childExited = false;

-child.once("exit", () => {
+child.once("exit", (code, signal) => {
  childExited = true;
+  if (pending.size > 0) {
+    const error = new Error(`MCP child exited before responding (code=${code}, signal=${signal}). stderr: ${stderrBuffer.slice(-2000)}`);
+    for (const { reject } of pending.values()) reject(error);
+    pending.clear();
+  }
 });

 function stopChild() {
  return new Promise(resolve => {
    if (childExited) {
+      stopContainer();
+      rmSync(tmpDir, { recursive: true, force: true });
      resolve();
      return;
    }

    child.stdin.end();
+    const stopTimer = setTimeout(() => {
+      stopContainer();
+    }, 1000);
    const termTimer = setTimeout(() => {
      if (!childExited) child.kill("SIGTERM");
    }, 3000);
@@ -39,13 +55,23 @@ function stopChild() {
    }, 6000);

    child.once("exit", () => {
+      stopContainer();
+      clearTimeout(stopTimer);
      clearTimeout(termTimer);
      clearTimeout(killTimer);
+      rmSync(tmpDir, { recursive: true, force: true });
      resolve();
    });
  });
 }

+function stopContainer() {
+  if (!existsSync(cidFile)) return;
+  const containerId = readFileSync(cidFile, "utf8").trim();
+  if (!containerId) return;
+  spawnSync("docker", ["stop", containerId], { stdio: "ignore" });
+}
+
 const timeout = setTimeout(async () => {
  await stopChild();
  console.error(`MCP smoke timed out. stderr: ${stderrBuffer.slice(-2000)}`);
@@ -79,6 +105,9 @@ child.stdout.on("data", chunk => {
 });

 function request(method, params = {}) {
+  if (childExited) {
+    return Promise.reject(new Error(`MCP child already exited. stderr: ${stderrBuffer.slice(-2000)}`));
+  }
  const id = nextId++;
  child.stdin.write(`${JSON.stringify({ jsonrpc: "2.0", id, method, params })}\n`);
  return new Promise((resolve, reject) => pending.set(id, { resolve, reject }));