feat(miner): add repo stack auto-detection (#4785)

packages/gittensory-miner/README.md

@@ -35,6 +35,14 @@ tenant goal spec to the ranker, printing `usedDefaultGoalSpec` so a fall-back to
 rather than silent. See [`docs/repo-agnostic-capability-audit.md`](docs/repo-agnostic-capability-audit.md) for the
 #4780 audit this executes.
 
+The package also includes repo stack auto-detection: `detectRepoStack` (`lib/stack-detection.js`) inspects an
+already-cloned target repo's manifest / lockfile / config files and returns a structured description — language,
+package manager, and the build / test / lint / format commands — for Node (npm/yarn/pnpm/bun), Python
+(pip/poetry/pipenv/uv), Rust, Go, Maven, and Gradle. It is pure (injectable `existsSync` / `readFileSync`), never
+throws, and per its acceptance criteria **fails closed** — a repo with no recognized manifest returns
+`{ detected: false, reason }` and a command that can't be inferred without guessing stays `null`, rather than being
+assumed. Detection only; wiring the description into the attempt prompt is the follow-up ([#4786](https://github.com/JSONbored/gittensory/issues/4786)). (#4785)
+
 The package also includes an append-only governor decision ledger: `initGovernorLedger` / `appendGovernorEvent`
 persist structured allow/deny/throttle/kill-switch outcomes in local SQLite for contributor audit. Insert-only —
 no enforcement wiring yet. (#2328)

packages/gittensory-miner/lib/stack-detection.d.ts

@@ -0,0 +1,41 @@
+/** Stack auto-detection (#4785). `detectRepoStack` inspects an already-cloned repo's manifest / lockfile / config
+ * files and returns a structured stack description, or an explicit fail-closed result when the stack can't be
+ * confidently identified (no guessing). */
+
+/** Which manifest (and lockfile, when present) drove the detection. */
+export type StackEvidence = {
+  manifest: string;
+  lockfile: string | null;
+};
+
+/** A confidently-detected stack. Command fields are `null` when the command can't be inferred without guessing. */
+export type DetectedRepoStack = {
+  detected: true;
+  language: string;
+  packageManager: string | null;
+  buildCommand: string | null;
+  testCommand: string | null;
+  lintCommand: string | null;
+  formatCommand: string | null;
+  evidence: StackEvidence;
+};
+
+/** A repo whose stack could not be confidently identified. */
+export type UndetectedRepoStack = {
+  detected: false;
+  reason: string;
+};
+
+export type RepoStackResult = DetectedRepoStack | UndetectedRepoStack;
+
+export type DetectRepoStackOptions = {
+  existsSync?: (path: string) => boolean;
+  readFileSync?: (path: string, encoding: "utf8") => string;
+};
+
+/** Manifests, in the precedence order detection tries them (first match wins). */
+export const RECOGNIZED_MANIFESTS: readonly string[];
+
+export function detectRepoStack(repoPath: string, options?: DetectRepoStackOptions): RepoStackResult;
+
+export function renderStackSummary(stack: RepoStackResult): string;

packages/gittensory-miner/lib/stack-detection.js

@@ -0,0 +1,248 @@
+/** Stack auto-detection (#4785): inspect an already-cloned target repo's manifest / lockfile / config files and
+ * infer a structured description of its stack — language, package manager, and the build / test / lint / format
+ * commands — before any code-generation step runs. Like `miner-goal-spec.js` this reads the ALREADY-CLONED repo on
+ * disk (attempt-worktree.js's prepareAttemptWorktree runs first), so the injected `existsSync` / `readFileSync`
+ * always receive the FULL joined path, mirroring node:fs. It is pure and NEVER throws: an unreadable/unparseable
+ * file degrades to "no evidence" rather than crashing, and — per the acceptance criteria — a repo whose stack
+ * can't be confidently identified returns an explicit `{ detected: false, reason }` instead of guessing. */
+import { existsSync, readFileSync } from "node:fs";
+import { join } from "node:path";
+
+/** Manifests, in the precedence order detection tries them; the first matching primary manifest wins. A caller with
+ * a known polyglot repo can inspect `evidence.manifest` to see which one was chosen. */
+export const RECOGNIZED_MANIFESTS = Object.freeze([
+  "package.json",
+  "pyproject.toml",
+  "setup.py",
+  "setup.cfg",
+  "requirements.txt",
+  "Pipfile",
+  "Cargo.toml",
+  "go.mod",
+  "pom.xml",
+  "build.gradle",
+  "build.gradle.kts",
+]);
+
+const NO_MANIFEST_REASON =
+  "No recognized dependency manifest (package.json, pyproject.toml, Cargo.toml, go.mod, pom.xml, or build.gradle) was found at the repository root.";
+
+const NODE_PACKAGE_MANAGERS = Object.freeze(["npm", "yarn", "pnpm", "bun"]);
+const NODE_LOCKFILES = Object.freeze([
+  ["pnpm-lock.yaml", "pnpm"],
+  ["yarn.lock", "yarn"],
+  ["bun.lockb", "bun"],
+  ["package-lock.json", "npm"],
+]);
+
+/** Build a never-throwing accessor over the cloned repo. `exists` and `read` both swallow fs errors so the detector
+ * treats an EACCES/ENOENT/binary file as simply "absent" instead of crashing the attempt. */
+function makeAccess(repoPath, options) {
+  const existsImpl = options.existsSync ?? existsSync;
+  const readImpl = options.readFileSync ?? readFileSync;
+  const exists = (relativePath) => {
+    try {
+      return existsImpl(join(repoPath, relativePath)) === true;
+    } catch {
+      return false;
+    }
+  };
+  const read = (relativePath) => {
+    try {
+      if (!exists(relativePath)) return null;
+      const content = readImpl(join(repoPath, relativePath), "utf8");
+      return typeof content === "string" ? content : null;
+    } catch {
+      return null;
+    }
+  };
+  return { exists, read };
+}
+
+function parseJson(text) {
+  if (typeof text !== "string") return null;
+  try {
+    const parsed = JSON.parse(text);
+    return parsed && typeof parsed === "object" ? parsed : null;
+  } catch {
+    return null;
+  }
+}
+
+/** Pick a package.json script by exact name first, then by pattern, considering only string-valued scripts. */
+function pickScript(scripts, exactName, pattern) {
+  const names = Object.keys(scripts).filter((name) => typeof scripts[name] === "string");
+  if (names.includes(exactName)) return exactName;
+  return names.find((name) => pattern.test(name)) ?? null;
+}
+
+function nodeLockfile(exists) {
+  const match = NODE_LOCKFILES.find(([file]) => exists(file));
+  return match ? match[0] : null;
+}
+
+function nodePackageManager(pkg, lockfile) {
+  const corepack =
+    typeof pkg?.packageManager === "string" ? pkg.packageManager.split("@")[0].trim().toLowerCase() : "";
+  if (NODE_PACKAGE_MANAGERS.includes(corepack)) return corepack;
+  const byLock = NODE_LOCKFILES.find(([file]) => file === lockfile);
+  // A package.json with no lockfile is still a Node project; npm is its default runner (a default, not a guess).
+  return byLock ? byLock[1] : "npm";
+}
+
+function hasTypescriptDependency(pkg) {
+  const deps = { ...(pkg?.dependencies ?? {}), ...(pkg?.devDependencies ?? {}) };
+  return typeof deps.typescript === "string";
+}
+
+function detectNode({ exists, read }) {
+  if (!exists("package.json")) return null;
+  const pkg = parseJson(read("package.json"));
+  const scripts =
+    pkg && typeof pkg.scripts === "object" && pkg.scripts && !Array.isArray(pkg.scripts) ? pkg.scripts : {};
+  const language = exists("tsconfig.json") || hasTypescriptDependency(pkg) ? "typescript" : "javascript";
+  const lockfile = nodeLockfile(exists);
+  const packageManager = nodePackageManager(pkg, lockfile);
+
+  const buildName = pickScript(scripts, "build", /^(build|compile|bundle)(:|$)/i);
+  const testName = pickScript(scripts, "test", /(^|:)test(:|$)/i);
+  const lintName = pickScript(scripts, "lint", /(^|:)lint(:|$)/i);
+  const formatName = pickScript(scripts, "format", /(^|:)(format|fmt)(:|$)/i);
+
+  return {
+    language,
+    packageManager,
+    buildCommand: buildName ? `${packageManager} run ${buildName}` : null,
+    // `<pm> test` is the built-in test lifecycle across npm/yarn/pnpm/bun; a non-"test" script uses `run`.
+    testCommand: testName ? (testName === "test" ? `${packageManager} test` : `${packageManager} run ${testName}`) : null,
+    lintCommand: lintName ? `${packageManager} run ${lintName}` : null,
+    formatCommand: formatName ? `${packageManager} run ${formatName}` : null,
+    evidence: { manifest: "package.json", lockfile },
+  };
+}
+
+function detectPython({ exists, read }) {
+  const manifest = ["pyproject.toml", "setup.py", "setup.cfg", "requirements.txt", "Pipfile"].find(exists);
+  if (manifest === undefined) return null;
+  const pyproject = read("pyproject.toml") ?? "";
+
+  let packageManager;
+  let lockfile = null;
+  if (exists("poetry.lock") || /\[tool\.poetry\]/.test(pyproject)) {
+    packageManager = "poetry";
+    lockfile = exists("poetry.lock") ? "poetry.lock" : null;
+  } else if (exists("uv.lock")) {
+    packageManager = "uv";
+    lockfile = "uv.lock";
+  } else if (exists("Pipfile") || exists("Pipfile.lock")) {
+    packageManager = "pipenv";
+    lockfile = exists("Pipfile.lock") ? "Pipfile.lock" : null;
+  } else {
+    packageManager = "pip";
+  }
+
+  // Commands are inferred only from real config so an undeclared tool is never guessed (acceptance: fail safe).
+  const hasRuff = exists("ruff.toml") || exists(".ruff.toml") || /\[tool\.ruff\]/.test(pyproject);
+  const hasPytest = exists("pytest.ini") || exists("tox.ini") || /\[tool\.pytest\b/.test(pyproject);
+
+  return {
+    language: "python",
+    packageManager,
+    buildCommand: /\[build-system\]/.test(pyproject) ? (packageManager === "poetry" ? "poetry build" : "python -m build") : null,
+    testCommand: hasPytest ? "pytest" : null,
+    lintCommand: hasRuff ? "ruff check ." : null,
+    formatCommand: hasRuff ? "ruff format ." : null,
+    evidence: { manifest, lockfile },
+  };
+}
+
+function detectRust({ exists }) {
+  if (!exists("Cargo.toml")) return null;
+  return {
+    language: "rust",
+    packageManager: "cargo",
+    buildCommand: "cargo build",
+    testCommand: "cargo test",
+    lintCommand: "cargo clippy",
+    formatCommand: "cargo fmt",
+    evidence: { manifest: "Cargo.toml", lockfile: exists("Cargo.lock") ? "Cargo.lock" : null },
+  };
+}
+
+function detectGo({ exists }) {
+  if (!exists("go.mod")) return null;
+  const hasGolangci = exists(".golangci.yml") || exists(".golangci.yaml") || exists(".golangci.toml");
+  return {
+    language: "go",
+    packageManager: "go",
+    buildCommand: "go build ./...",
+    testCommand: "go test ./...",
+    lintCommand: hasGolangci ? "golangci-lint run" : "go vet ./...",
+    formatCommand: "gofmt -l .",
+    evidence: { manifest: "go.mod", lockfile: exists("go.sum") ? "go.sum" : null },
+  };
+}
+
+function detectMaven({ exists }) {
+  if (!exists("pom.xml")) return null;
+  return {
+    language: "java",
+    packageManager: "maven",
+    buildCommand: "mvn -B package",
+    testCommand: "mvn -B test",
+    lintCommand: null,
+    formatCommand: null,
+    evidence: { manifest: "pom.xml", lockfile: null },
+  };
+}
+
+function detectGradle({ exists }) {
+  const manifest = exists("build.gradle") ? "build.gradle" : exists("build.gradle.kts") ? "build.gradle.kts" : null;
+  if (manifest === null) return null;
+  const runner = exists("gradlew") ? "./gradlew" : "gradle";
+  return {
+    language: "java",
+    packageManager: "gradle",
+    buildCommand: `${runner} build`,
+    testCommand: `${runner} test`,
+    lintCommand: null,
+    formatCommand: null,
+    evidence: { manifest, lockfile: null },
+  };
+}
+
+const DETECTORS = Object.freeze([detectNode, detectPython, detectRust, detectGo, detectMaven, detectGradle]);
+
+/**
+ * Detect the stack of an already-cloned repository at `repoPath`. Returns `{ detected: true, ... }` with the
+ * language, package manager, and any confidently-inferred commands, or `{ detected: false, reason }` when no
+ * recognized manifest is present. Never throws.
+ */
+export function detectRepoStack(repoPath, options = {}) {
+  if (typeof repoPath !== "string" || !repoPath.trim()) {
+    return { detected: false, reason: "A repository path is required to detect the stack." };
+  }
+  const access = makeAccess(repoPath, options);
+  for (const detector of DETECTORS) {
+    const detected = detector(access);
+    if (detected !== null) {
+      return { detected: true, ...detected };
+    }
+  }
+  return { detected: false, reason: NO_MANIFEST_REASON };
+}
+
+/** One-line human summary of a detection result, suitable for a coding-agent prompt or an operator log. */
+export function renderStackSummary(stack) {
+  if (!stack || stack.detected !== true) {
+    return `stack not detected: ${stack?.reason ?? "unknown reason"}`;
+  }
+  const commands = [
+    stack.buildCommand ? `build=\`${stack.buildCommand}\`` : null,
+    stack.testCommand ? `test=\`${stack.testCommand}\`` : null,
+    stack.lintCommand ? `lint=\`${stack.lintCommand}\`` : null,
+    stack.formatCommand ? `format=\`${stack.formatCommand}\`` : null,
+  ].filter((entry) => entry !== null);
+  const suffix = commands.length > 0 ? ` (${commands.join(", ")})` : " (no validation commands detected)";
+  return `${stack.language} via ${stack.packageManager ?? "unknown"}${suffix}`;
+}

packages/gittensory-miner/package.json

@@ -33,7 +33,7 @@
     "expected-engine.version"
   ],
   "scripts": {
-    "build": "node --check bin/gittensory-miner.js && node --check bin/gittensory-miner-mcp.js && node --check lib/ams-policy.js && node --check lib/attempt-cli.js && node --check lib/attempt-input-builder.js && node --check lib/attempt-log.js && node --check lib/attempt-runner.js && node --check lib/attempt-worktree.js && node --check lib/calibration-run.js && node --check lib/calibration-types.js && node --check lib/calibration.js && node --check lib/ci-poller.js && node --check lib/claim-adjudication.js && node --check lib/claim-ledger-cli.js && node --check lib/claim-ledger-expiry.js && node --check lib/claim-ledger.js && node --check lib/cli.js && node --check lib/coding-agent-construction.js && node --check lib/coding-agent-house-rules.js && node --check lib/coding-task-spec.js && node --check lib/deny-check.js && node --check lib/deny-hook-synthesis.js && node --check lib/deny-hooks.js && node --check lib/deployment-docs-audit.js && node --check lib/discover-cli.js && node --check lib/event-ledger-cli.js && node --check lib/event-ledger.js && node --check lib/execute-local-write.js && node --check lib/feasibility-cli.js && node --check lib/gate-verdict-poller.js && node --check lib/governor-action-mode.js && node --check lib/governor-chokepoint-persisted.js && node --check lib/governor-chokepoint.js && node --check lib/governor-kill-switch.js && node --check lib/governor-ledger-cli.js && node --check lib/governor-ledger.js && node --check lib/governor-open-pr.js && node --check lib/governor-run-halt.js && node --check lib/governor-state.js && node --check lib/governor-write-rate-limit.js && node --check lib/harness-submission-trigger.js && node --check lib/laptop-init.js && node --check lib/live-issue-snapshot.js && node --check lib/local-store.js && node --check lib/loop-cli.js && node --check lib/loop-closure.js && node --check lib/loop-reentry.js && node --check lib/manage-poll.js && node --check lib/manage-status.js && node --check lib/metrics-cli.js && node --check lib/miner-goal-spec.js && node --check lib/opportunity-fanout.js && node --check lib/opportunity-ranker.js && node --check lib/orb-export.js && node --check lib/plan-store-cli.js && node --check lib/plan-store.js && node --check lib/portfolio-dashboard.js && node --check lib/portfolio-discovery.js && node --check lib/portfolio-queue-cli.js && node --check lib/portfolio-queue-manager.js && node --check lib/portfolio-queue.js && node --check lib/portfolio-queue-expiry.js && node --check lib/pr-disposition-poller.js && node --check lib/pr-outcome.js && node --check lib/prediction-ledger.js && node --check lib/pretooluse-hook.js && node --check lib/rejection-signal.js && node --check lib/rejection-state-machine.js && node --check lib/rejection-templates.js && node --check lib/replay-objective-anchor.js && node --check lib/replay-snapshot.js && node --check lib/replay-task-generation.js && node --check lib/repo-clone.js && node --check lib/run-state-cli.js && node --check lib/run-state.js && node --check lib/self-review-context.js && node --check lib/slop-assessment.js && node --check lib/status.js && node --check lib/submission-freshness-check.js && node --check lib/update-check.js && node --check lib/version.js && node --check lib/worktree-allocator.js"
+    "build": "node --check bin/gittensory-miner.js && node --check bin/gittensory-miner-mcp.js && node --check lib/ams-policy.js && node --check lib/attempt-cli.js && node --check lib/attempt-input-builder.js && node --check lib/attempt-log.js && node --check lib/attempt-runner.js && node --check lib/attempt-worktree.js && node --check lib/calibration-run.js && node --check lib/calibration-types.js && node --check lib/calibration.js && node --check lib/ci-poller.js && node --check lib/claim-adjudication.js && node --check lib/claim-ledger-cli.js && node --check lib/claim-ledger-expiry.js && node --check lib/claim-ledger.js && node --check lib/cli.js && node --check lib/coding-agent-construction.js && node --check lib/coding-agent-house-rules.js && node --check lib/coding-task-spec.js && node --check lib/deny-check.js && node --check lib/deny-hook-synthesis.js && node --check lib/deny-hooks.js && node --check lib/deployment-docs-audit.js && node --check lib/discover-cli.js && node --check lib/event-ledger-cli.js && node --check lib/event-ledger.js && node --check lib/execute-local-write.js && node --check lib/feasibility-cli.js && node --check lib/gate-verdict-poller.js && node --check lib/governor-action-mode.js && node --check lib/governor-chokepoint-persisted.js && node --check lib/governor-chokepoint.js && node --check lib/governor-kill-switch.js && node --check lib/governor-ledger-cli.js && node --check lib/governor-ledger.js && node --check lib/governor-open-pr.js && node --check lib/governor-run-halt.js && node --check lib/governor-state.js && node --check lib/governor-write-rate-limit.js && node --check lib/harness-submission-trigger.js && node --check lib/laptop-init.js && node --check lib/live-issue-snapshot.js && node --check lib/local-store.js && node --check lib/loop-cli.js && node --check lib/loop-closure.js && node --check lib/loop-reentry.js && node --check lib/manage-poll.js && node --check lib/manage-status.js && node --check lib/metrics-cli.js && node --check lib/miner-goal-spec.js && node --check lib/opportunity-fanout.js && node --check lib/opportunity-ranker.js && node --check lib/orb-export.js && node --check lib/plan-store-cli.js && node --check lib/plan-store.js && node --check lib/portfolio-dashboard.js && node --check lib/portfolio-discovery.js && node --check lib/portfolio-queue-cli.js && node --check lib/portfolio-queue-manager.js && node --check lib/portfolio-queue.js && node --check lib/portfolio-queue-expiry.js && node --check lib/pr-disposition-poller.js && node --check lib/pr-outcome.js && node --check lib/prediction-ledger.js && node --check lib/pretooluse-hook.js && node --check lib/rejection-signal.js && node --check lib/rejection-state-machine.js && node --check lib/rejection-templates.js && node --check lib/replay-objective-anchor.js && node --check lib/replay-snapshot.js && node --check lib/replay-task-generation.js && node --check lib/repo-clone.js && node --check lib/run-state-cli.js && node --check lib/run-state.js && node --check lib/self-review-context.js && node --check lib/slop-assessment.js && node --check lib/stack-detection.js && node --check lib/status.js && node --check lib/submission-freshness-check.js && node --check lib/update-check.js && node --check lib/version.js && node --check lib/worktree-allocator.js"
   },
   "dependencies": {
     "@jsonbored/gittensory-engine": "*",