huskies/server/src/agents/gates.rs

//! Acceptance gates — runs test suites and validation scripts in agent worktrees.
use serde::{Deserialize, Serialize};
use std::path::Path;
use std::process::Command;
use std::time::Duration;
use wait_timeout::ChildExt;

/// Typed classification of a gate failure, produced at the gate execution boundary.
///
/// Downstream decision logic (e.g. `is_self_evident_fix`) matches on the variant
/// rather than scanning the raw output string for patterns.
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
pub enum GateFailureKind {
    /// `cargo fmt --check` or `rustfmt --check` detected formatting drift.
    Fmt,
    /// `cargo clippy` produced warnings or errors (promoted via `-D warnings`).
    Lint,
    /// Test suite (`script/test`, `cargo nextest`, `cargo test`) failed.
    Test,
    /// `source-map-check` gate found missing or incomplete doc comments.
    SourceMapCheck,
    /// Git content conflict detected during squash-rebase.
    ContentConflict,
    /// Build-level failure (duplicate module files E0761, compile error).
    Build,
    /// Unclassified gate failure.
    Other,
}

impl GateFailureKind {
    /// Classify a gate failure from its raw output at the gate execution boundary.
    ///
    /// Called once when a gate fails to produce a typed kind.  Downstream code
    /// matches on the variant and must not call this on subsequent reads.
    pub fn classify(output: &str) -> Self {
        // Strip `test <name> ... ok` lines before checking lint-trigger keywords so
        // a passing test whose name contains e.g. `missing_doc_comments` or `clippy::`
        // does not produce a false-positive Lint classification (story 1101).
        let stripped_for_lint: String = output
            .lines()
            .filter(|l| {
                let t = l.trim();
                !(t.starts_with("test ") && t.ends_with("... ok"))
            })
            .collect::<Vec<_>>()
            .join("\n");
        let is_lint = stripped_for_lint.contains("error[clippy::")
            || stripped_for_lint.contains("warning[clippy::")
            || stripped_for_lint.contains("missing_doc_comments");

        if output.contains("CONFLICT (content):") || output.contains("Merge conflict:") {
            GateFailureKind::ContentConflict
        } else if output.contains("Diff in ") || output.contains("would reformat") {
            GateFailureKind::Fmt
        } else if output.contains("missing-docs direction") {
            GateFailureKind::SourceMapCheck
        } else if is_lint {
            GateFailureKind::Lint
        } else if output.contains("error[E") {
            // rustc compile errors (e.g. `error[E0063]: missing field`).
            // When this appears in the post-squash gate run, it almost always
            // signals cross-merge breakage — master gained a field/variant the
            // feature branch's code does not match. Mergemaster handles the
            // recovery in its ConflictDetected path.
            GateFailureKind::Build
        } else {
            GateFailureKind::Test
        }
    }

    /// Whether this failure class is a self-evident fix that a short coder session
    /// can resolve without human intervention (fmt drift, lint warnings, missing docs).
    pub fn is_self_evident_fix(&self) -> bool {
        matches!(
            self,
            GateFailureKind::Fmt | GateFailureKind::Lint | GateFailureKind::SourceMapCheck
        )
    }
}

/// Outcome of running quality gates, produced at the gate execution boundary.
///
/// `failure_kind` drives routing decisions; `output` is retained for human-readable
/// display and injection into agent retry prompts only — it must not be used as a
/// decision source (story 986).
#[derive(Debug, Clone)]
pub struct GateOutcome {
    /// Whether all gates passed.
    pub passed: bool,
    /// Typed failure classification; `None` when `passed` is `true`.
    pub failure_kind: Option<GateFailureKind>,
    /// Human-readable combined gate output (display/prompt injection only).
    pub output: String,
}

impl GateOutcome {
    /// Passing outcome.
    pub(crate) fn pass(output: String) -> Self {
        Self {
            passed: true,
            failure_kind: None,
            output,
        }
    }

    /// Failing outcome — classifies `failure_kind` from the output at construction.
    pub(crate) fn fail(output: String) -> Self {
        let failure_kind = Some(GateFailureKind::classify(&output));
        Self {
            passed: false,
            failure_kind,
            output,
        }
    }

    /// Failing outcome for a pre-classified build error (e.g. duplicate module files).
    pub(crate) fn build_error(output: String) -> Self {
        Self {
            passed: false,
            failure_kind: Some(GateFailureKind::Build),
            output,
        }
    }
}

/// Maximum time any single test command is allowed to run before being killed.
const TEST_TIMEOUT: Duration = Duration::from_secs(1200); // 20 minutes

/// Detect whether the base branch in a worktree is `master` or `main`.
/// Falls back to `"master"` if neither is found.
pub(crate) fn detect_worktree_base_branch(wt_path: &Path) -> String {
    for branch in &["master", "main"] {
        let ok = Command::new("git")
            .args(["rev-parse", "--verify", branch])
            .current_dir(wt_path)
            .output()
            .map(|o| o.status.success())
            .unwrap_or(false);
        if ok {
            return branch.to_string();
        }
    }
    "master".to_string()
}

/// Return `true` if the git worktree at `wt_path` has commits on its current
/// branch that are not present on the base branch (`master` or `main`).
///
/// Used during server startup reconciliation to detect stories whose agent work
/// was committed while the server was offline.
pub(crate) fn worktree_has_committed_work(wt_path: &Path) -> bool {
    let base_branch = detect_worktree_base_branch(wt_path);
    let output = Command::new("git")
        .args(["log", &format!("{base_branch}..HEAD"), "--oneline"])
        .current_dir(wt_path)
        .output();
    match output {
        Ok(out) if out.status.success() => !String::from_utf8_lossy(&out.stdout).trim().is_empty(),
        _ => false,
    }
}

/// Run `cargo check` in the given worktree directory to verify committed code compiles.
///
/// Stashes any dirty (uncommitted) files first so that only the committed state
/// is evaluated, then restores them afterward.  Uncommitted work in worktrees is
/// never junk — it may be the next agent session's starting point (bug 651).
///
/// No longer called from main pipeline code (bug 668 replaced cargo-check with
/// run_tests evidence), but retained for the bug-651 stash/restore regression test.
#[cfg(test)]
pub(crate) fn cargo_check_in_worktree(wt_path: &Path) -> bool {
    // Stash uncommitted changes (including untracked files) so cargo check
    // evaluates only committed code.  We restore them afterward.
    let stashed = Command::new("git")
        .args([
            "stash",
            "push",
            "--include-untracked",
            "-m",
            "cargo-check-temp",
        ])
        .current_dir(wt_path)
        .output()
        .map(|o| {
            o.status.success()
                && !String::from_utf8_lossy(&o.stdout).contains("No local changes to save")
        })
        .unwrap_or(false);

    let result = Command::new("cargo")
        .args(["check"])
        .current_dir(wt_path)
        .output()
        .map(|o| o.status.success())
        .unwrap_or(false);

    // Restore stashed uncommitted changes.
    if stashed {
        let _ = Command::new("git")
            .args(["stash", "pop"])
            .current_dir(wt_path)
            .output();
    }

    result
}

/// Check whether the given directory has any uncommitted git changes.
/// Returns `Err` with a descriptive message if there are any.
pub(crate) fn check_uncommitted_changes(path: &Path) -> Result<(), String> {
    let output = Command::new("git")
        .args(["status", "--porcelain"])
        .current_dir(path)
        .output()
        .map_err(|e| format!("Failed to run git status: {e}"))?;

    let stdout = String::from_utf8_lossy(&output.stdout);
    if !stdout.trim().is_empty() {
        return Err(format!(
            "Worktree has uncommitted changes. Please commit all work before \
             the agent exits:\n{stdout}"
        ));
    }
    Ok(())
}

/// Run the project's test suite.
///
/// Uses `script/test` if present, treating it as the canonical single test entry point.
/// Falls back to `cargo nextest run` / `cargo test` when `script/test` is absent.
/// Returns `(tests_passed, output)`.
pub(crate) fn run_project_tests(path: &Path) -> Result<(bool, String), String> {
    let script_test = path.join("script").join("test");
    if script_test.exists() {
        let mut output = String::from("=== script/test ===\n");
        let (success, out) = run_command_with_timeout(&script_test, &[], path)?;
        output.push_str(&out);
        output.push('\n');
        return Ok((success, output));
    }

    // Fallback: cargo nextest run / cargo test
    let mut output = String::from("=== tests ===\n");
    let (success, test_out) = match run_command_with_timeout("cargo", &["nextest", "run"], path) {
        Ok(result) => result,
        Err(_) => {
            // nextest not available — fall back to cargo test
            run_command_with_timeout("cargo", &["test"], path)
                .map_err(|e| format!("Failed to run cargo test: {e}"))?
        }
    };
    output.push_str(&test_out);
    output.push('\n');
    Ok((success, output))
}

/// Run a command with a timeout. Returns `(success, combined_output)`.
/// Kills the child process if it exceeds `TEST_TIMEOUT`.
///
/// Stdout and stderr are drained in background threads to avoid a pipe-buffer
/// deadlock: if the child fills the 64 KB OS pipe buffer while the parent
/// blocks on `waitpid`, neither side can make progress.
fn run_command_with_timeout(
    program: impl AsRef<std::ffi::OsStr>,
    args: &[&str],
    dir: &Path,
) -> Result<(bool, String), String> {
    // On Linux, execve can return ETXTBSY (26) briefly after a file is written
    // before the kernel releases its "write open" state. Retry once after a
    // short pause to handle this race condition.
    let mut last_err = None;
    let mut cmd = Command::new(&program);
    cmd.args(args)
        .current_dir(dir)
        .stdout(std::process::Stdio::piped())
        .stderr(std::process::Stdio::piped());
    let mut child = loop {
        match cmd.spawn() {
            Ok(c) => break c,
            Err(e) if e.raw_os_error() == Some(26) => {
                // ETXTBSY — wait briefly and retry once
                if last_err.is_some() {
                    return Err(format!("Failed to spawn command: {e}"));
                }
                last_err = Some(e);
                std::thread::sleep(std::time::Duration::from_millis(50));
            }
            Err(e) => return Err(format!("Failed to spawn command: {e}")),
        }
    };

    // Drain stdout/stderr in background threads so the pipe buffers never fill.
    let stdout_handle = child.stdout.take().map(|r| {
        std::thread::spawn(move || {
            let mut s = String::new();
            let mut r = r;
            std::io::Read::read_to_string(&mut r, &mut s).ok();
            s
        })
    });
    let stderr_handle = child.stderr.take().map(|r| {
        std::thread::spawn(move || {
            let mut s = String::new();
            let mut r = r;
            std::io::Read::read_to_string(&mut r, &mut s).ok();
            s
        })
    });

    match child.wait_timeout(TEST_TIMEOUT) {
        Ok(Some(status)) => {
            let stdout = stdout_handle
                .and_then(|h| h.join().ok())
                .unwrap_or_default();
            let stderr = stderr_handle
                .and_then(|h| h.join().ok())
                .unwrap_or_default();
            Ok((status.success(), format!("{stdout}{stderr}")))
        }
        Ok(None) => {
            // Timed out — kill the child.
            let _ = child.kill();
            let _ = child.wait();
            Err(format!(
                "Command timed out after {} seconds",
                TEST_TIMEOUT.as_secs()
            ))
        }
        Err(e) => Err(format!("Failed to wait for command: {e}")),
    }
}

/// Run `cargo clippy` and the project test suite (via `script/test` if present,
/// otherwise `cargo nextest run` / `cargo test`) in the given directory.
/// Returns a [`GateOutcome`] with a typed failure classification.
pub(crate) fn run_acceptance_gates(path: &Path) -> Result<GateOutcome, String> {
    // Pre-flight: detect duplicate module files (E0761) before running the
    // full test suite so the failure message is immediately actionable.
    let duplicates = find_duplicate_module_files(path);
    if !duplicates.is_empty() {
        let mut msg = String::from(
            "ERROR [E0761]: duplicate module files detected — cargo will fail to compile.\n\
             Fix: git rm the flat .rs file in the same commit that introduces the mod.rs.\n",
        );
        for (flat, mod_path) in &duplicates {
            msg.push_str(&format!(
                "  {} (conflicts with {})\n",
                flat.display(),
                mod_path.display()
            ));
        }
        return Ok(GateOutcome::build_error(msg));
    }

    // Run script/test (or fallback to cargo test). Project-specific linting
    // and test commands belong in script/test.
    let (test_success, test_out) = run_project_tests(path)?;
    if !test_success {
        return Ok(GateOutcome::fail(test_out));
    }

    Ok(GateOutcome::pass(test_out))
}

/// Scan `root` recursively for Rust source files where both `path/X.rs` and
/// `path/X/mod.rs` exist — a condition that produces a `duplicate module file`
/// cargo error (E0761).
///
/// Returns a sorted list of `(flat_path, mod_path)` pairs in conflict.
/// Directories named `target` or starting with `.` are skipped.
pub(crate) fn find_duplicate_module_files(
    root: &Path,
) -> Vec<(std::path::PathBuf, std::path::PathBuf)> {
    let mut duplicates = Vec::new();
    find_duplicates_recursive(root, &mut duplicates);
    duplicates.sort();
    duplicates
}

fn find_duplicates_recursive(dir: &Path, out: &mut Vec<(std::path::PathBuf, std::path::PathBuf)>) {
    let entries = match std::fs::read_dir(dir) {
        Ok(e) => e,
        Err(_) => return,
    };
    for entry in entries.flatten() {
        let path = entry.path();
        if !path.is_dir() {
            continue;
        }
        let name = match path.file_name().and_then(|n| n.to_str()) {
            Some(n) => n.to_owned(),
            None => continue,
        };
        if name == "target" || name.starts_with('.') {
            continue;
        }
        let flat = dir.join(format!("{name}.rs"));
        let mod_rs = path.join("mod.rs");
        if flat.exists() && mod_rs.exists() {
            out.push((flat, mod_rs));
        }
        find_duplicates_recursive(&path, out);
    }
}

/// Run `script/test_coverage` in the given directory if the script exists.
///
/// Used as a QA gate before advancing a story from `3_qa/` to `4_merge/`.
/// Returns `(passed, output)`.  If the script does not exist, returns `(true, …)`.
pub(crate) fn run_coverage_gate(path: &Path) -> Result<(bool, String), String> {
    let script = path.join("script").join("test_coverage");
    if !script.exists() {
        return Ok((
            true,
            "script/test_coverage not found; coverage gate skipped.\n".to_string(),
        ));
    }

    let mut output = String::from("=== script/test_coverage ===\n");
    let result = match Command::new(&script).current_dir(path).output() {
        Ok(r) => r,
        Err(e) if e.raw_os_error() == Some(26) => {
            // ETXTBSY — retry once after a brief pause.
            std::thread::sleep(std::time::Duration::from_millis(50));
            Command::new(&script)
                .current_dir(path)
                .output()
                .map_err(|e| format!("Failed to run script/test_coverage: {e}"))?
        }
        Err(e) => return Err(format!("Failed to run script/test_coverage: {e}")),
    };

    let combined = format!(
        "{}{}",
        String::from_utf8_lossy(&result.stdout),
        String::from_utf8_lossy(&result.stderr)
    );
    output.push_str(&combined);
    output.push('\n');

    Ok((result.status.success(), output))
}

#[cfg(test)]
mod tests {
    use super::*;

    fn init_git_repo(repo: &std::path::Path) {
        Command::new("git")
            .args(["init"])
            .current_dir(repo)
            .output()
            .unwrap();
        Command::new("git")
            .args(["config", "user.email", "test@test.com"])
            .current_dir(repo)
            .output()
            .unwrap();
        Command::new("git")
            .args(["config", "user.name", "Test"])
            .current_dir(repo)
            .output()
            .unwrap();
        Command::new("git")
            .args(["commit", "--allow-empty", "-m", "init"])
            .current_dir(repo)
            .output()
            .unwrap();
    }

    // ── find_duplicate_module_files tests ────────────────────────

    #[test]
    fn find_duplicate_module_files_returns_empty_when_no_duplicates() {
        use std::fs;
        let tmp = tempfile::tempdir().unwrap();
        let src = tmp.path().join("src");
        fs::create_dir_all(&src).unwrap();
        // Only X/mod.rs, no X.rs
        let sub = src.join("util");
        fs::create_dir_all(&sub).unwrap();
        fs::write(sub.join("mod.rs"), "").unwrap();
        let result = find_duplicate_module_files(tmp.path());
        assert!(result.is_empty(), "no duplicates expected: {result:?}");
    }

    #[test]
    fn find_duplicate_module_files_detects_flat_and_mod_rs() {
        use std::fs;
        let tmp = tempfile::tempdir().unwrap();
        let src = tmp.path().join("src");
        fs::create_dir_all(&src).unwrap();
        // Create both src/config.rs and src/config/mod.rs
        fs::write(src.join("config.rs"), "").unwrap();
        let config_dir = src.join("config");
        fs::create_dir_all(&config_dir).unwrap();
        fs::write(config_dir.join("mod.rs"), "").unwrap();
        let result = find_duplicate_module_files(tmp.path());
        assert_eq!(
            result.len(),
            1,
            "expected exactly one duplicate: {result:?}"
        );
        let (flat, mod_path) = &result[0];
        assert!(
            flat.ends_with("src/config.rs"),
            "flat path should be config.rs, got {flat:?}"
        );
        assert!(
            mod_path.ends_with("src/config/mod.rs"),
            "mod path should be config/mod.rs, got {mod_path:?}"
        );
    }

    #[test]
    fn find_duplicate_module_files_skips_target_directory() {
        use std::fs;
        let tmp = tempfile::tempdir().unwrap();
        // Duplicate inside target/ should be ignored
        let target = tmp.path().join("target").join("debug").join("foo");
        fs::create_dir_all(&target).unwrap();
        fs::write(target.join("mod.rs"), "").unwrap();
        fs::write(tmp.path().join("target").join("debug").join("foo.rs"), "").unwrap();
        let result = find_duplicate_module_files(tmp.path());
        assert!(
            result.is_empty(),
            "duplicates inside target/ should be ignored: {result:?}"
        );
    }

    #[test]
    fn find_duplicate_module_files_reports_both_paths_in_message() {
        use std::fs;
        let tmp = tempfile::tempdir().unwrap();
        let src = tmp.path().join("src");
        fs::create_dir_all(&src).unwrap();
        fs::write(src.join("handlers.rs"), "").unwrap();
        let handlers_dir = src.join("handlers");
        fs::create_dir_all(&handlers_dir).unwrap();
        fs::write(handlers_dir.join("mod.rs"), "").unwrap();
        let result = find_duplicate_module_files(tmp.path());
        assert_eq!(result.len(), 1);
        let (flat, mod_path) = &result[0];
        // Both file names must appear in the paths so a caller can surface them
        assert!(flat.to_string_lossy().contains("handlers.rs"));
        assert!(mod_path.to_string_lossy().contains("handlers"));
        assert!(mod_path.to_string_lossy().contains("mod.rs"));
    }

    // ── run_project_tests tests ───────────────────────────────────

    #[cfg(unix)]
    #[test]
    fn run_project_tests_uses_script_test_when_present_and_passes() {
        use std::fs;
        use std::os::unix::fs::PermissionsExt;

        let tmp = tempfile::tempdir().unwrap();
        let path = tmp.path();
        let script_dir = path.join("script");
        fs::create_dir_all(&script_dir).unwrap();
        let script_test = script_dir.join("test");
        fs::write(
            &script_test,
            "#!/usr/bin/env bash\necho 'all tests passed'\nexit 0\n",
        )
        .unwrap();
        let mut perms = fs::metadata(&script_test).unwrap().permissions();
        perms.set_mode(0o755);
        fs::set_permissions(&script_test, perms).unwrap();

        let (passed, output) = run_project_tests(path).unwrap();
        assert!(passed, "script/test exiting 0 should pass");
        assert!(
            output.contains("script/test"),
            "output should mention script/test"
        );
    }

    #[cfg(unix)]
    #[test]
    fn run_project_tests_reports_failure_when_script_test_exits_nonzero() {
        use std::fs;
        use std::os::unix::fs::PermissionsExt;

        let tmp = tempfile::tempdir().unwrap();
        let path = tmp.path();
        let script_dir = path.join("script");
        fs::create_dir_all(&script_dir).unwrap();
        let script_test = script_dir.join("test");
        fs::write(&script_test, "#!/usr/bin/env bash\nexit 1\n").unwrap();
        let mut perms = fs::metadata(&script_test).unwrap().permissions();
        perms.set_mode(0o755);
        fs::set_permissions(&script_test, perms).unwrap();

        let (passed, output) = run_project_tests(path).unwrap();
        assert!(!passed, "script/test exiting 1 should fail");
        assert!(
            output.contains("script/test"),
            "output should mention script/test"
        );
    }

    // ── run_coverage_gate tests ───────────────────────────────────────────────

    #[cfg(unix)]
    #[test]
    fn coverage_gate_passes_when_script_absent() {
        use tempfile::tempdir;
        let tmp = tempdir().unwrap();
        let (passed, output) = run_coverage_gate(tmp.path()).unwrap();
        assert!(passed, "coverage gate should pass when script is absent");
        assert!(
            output.contains("not found"),
            "output should mention script not found"
        );
    }

    #[cfg(unix)]
    #[test]
    fn coverage_gate_passes_when_script_exits_zero() {
        use std::fs;
        use std::io::Write;
        use std::os::unix::fs::PermissionsExt;

        let tmp = tempfile::tempdir().unwrap();
        let path = tmp.path();
        let script_dir = path.join("script");
        fs::create_dir_all(&script_dir).unwrap();
        let script = script_dir.join("test_coverage");
        {
            let mut f = fs::File::create(&script).unwrap();
            f.write_all(b"#!/usr/bin/env bash\necho 'Rust line coverage: 85%'\necho 'PASS: Coverage 85% meets threshold 0%'\nexit 0\n").unwrap();
            f.sync_all().unwrap();
        }
        let mut perms = fs::metadata(&script).unwrap().permissions();
        perms.set_mode(0o755);
        fs::set_permissions(&script, perms).unwrap();

        let (passed, output) = run_coverage_gate(path).unwrap();
        assert!(passed, "coverage gate should pass when script exits 0");
        assert!(
            output.contains("script/test_coverage"),
            "output should mention script/test_coverage"
        );
    }

    #[cfg(unix)]
    #[test]
    fn coverage_gate_fails_when_script_exits_nonzero() {
        use std::fs;
        use std::io::Write;
        use std::os::unix::fs::PermissionsExt;

        let tmp = tempfile::tempdir().unwrap();
        let path = tmp.path();
        let script_dir = path.join("script");
        fs::create_dir_all(&script_dir).unwrap();
        let script = script_dir.join("test_coverage");
        {
            let mut f = fs::File::create(&script).unwrap();
            f.write_all(
                b"#!/usr/bin/env bash\necho 'FAIL: Coverage 40% is below threshold 80%'\nexit 1\n",
            )
            .unwrap();
            f.sync_all().unwrap();
        }
        let mut perms = fs::metadata(&script).unwrap().permissions();
        perms.set_mode(0o755);
        fs::set_permissions(&script, perms).unwrap();

        let (passed, output) = run_coverage_gate(path).unwrap();
        assert!(!passed, "coverage gate should fail when script exits 1");
        assert!(
            output.contains("script/test_coverage"),
            "output should mention script/test_coverage"
        );
    }

    // ── worktree_has_committed_work tests ─────────────────────────────────────

    #[test]
    fn worktree_has_committed_work_false_on_fresh_repo() {
        let tmp = tempfile::tempdir().unwrap();
        let repo = tmp.path();
        // init_git_repo creates the initial commit on the default branch.
        // HEAD IS the base branch — no commits ahead.
        init_git_repo(repo);
        assert!(!worktree_has_committed_work(repo));
    }

    #[test]
    fn worktree_has_committed_work_true_after_commit_on_feature_branch() {
        use std::fs;
        let tmp = tempfile::tempdir().unwrap();
        let project_root = tmp.path().join("project");
        fs::create_dir_all(&project_root).unwrap();
        init_git_repo(&project_root);

        // Create a git worktree on a feature branch.
        let wt_path = tmp.path().join("wt");
        Command::new("git")
            .args([
                "worktree",
                "add",
                &wt_path.to_string_lossy(),
                "-b",
                "feature/story-99_test",
            ])
            .current_dir(&project_root)
            .output()
            .unwrap();

        // No commits on the feature branch yet — same as base branch.
        assert!(!worktree_has_committed_work(&wt_path));

        // Add a commit to the feature branch in the worktree.
        fs::write(wt_path.join("work.txt"), "done").unwrap();
        Command::new("git")
            .args(["add", "."])
            .current_dir(&wt_path)
            .output()
            .unwrap();
        Command::new("git")
            .args([
                "-c",
                "user.email=test@test.com",
                "-c",
                "user.name=Test",
                "commit",
                "-m",
                "coder: implement story",
            ])
            .current_dir(&wt_path)
            .output()
            .unwrap();

        // Now the feature branch is ahead of the base branch.
        assert!(worktree_has_committed_work(&wt_path));
    }

    // ── cargo_check_in_worktree tests ────────────────────────────────────────

    /// Bug 645 + 651: cargo_check_in_worktree stashes dirty files before
    /// checking committed code and restores them afterward.
    #[test]
    fn cargo_check_in_worktree_stashes_and_restores_dirty_files() {
        use std::fs;
        let tmp = tempfile::tempdir().unwrap();
        let project_root = tmp.path().join("project");
        fs::create_dir_all(&project_root).unwrap();
        init_git_repo(&project_root);

        // Create a minimal Cargo project so cargo check works.
        fs::write(
            project_root.join("Cargo.toml"),
            "[package]\nname = \"test_proj\"\nversion = \"0.1.0\"\nedition = \"2021\"\n",
        )
        .unwrap();
        fs::create_dir_all(project_root.join("src")).unwrap();
        fs::write(project_root.join("src/lib.rs"), "// empty lib\n").unwrap();
        Command::new("git")
            .args(["add", "."])
            .current_dir(&project_root)
            .output()
            .unwrap();
        Command::new("git")
            .args([
                "-c",
                "user.email=test@test.com",
                "-c",
                "user.name=Test",
                "commit",
                "-m",
                "add cargo project",
            ])
            .current_dir(&project_root)
            .output()
            .unwrap();

        // Create a worktree on a feature branch.
        let wt_path = tmp.path().join("wt");
        Command::new("git")
            .args([
                "worktree",
                "add",
                &wt_path.to_string_lossy(),
                "-b",
                "feature/story-645_test",
            ])
            .current_dir(&project_root)
            .output()
            .unwrap();

        // Commit valid code.
        fs::write(wt_path.join("src/lib.rs"), "pub fn hello() {}\n").unwrap();
        Command::new("git")
            .args(["add", "."])
            .current_dir(&wt_path)
            .output()
            .unwrap();
        Command::new("git")
            .args([
                "-c",
                "user.email=test@test.com",
                "-c",
                "user.name=Test",
                "commit",
                "-m",
                "add hello fn",
            ])
            .current_dir(&wt_path)
            .output()
            .unwrap();

        // Now simulate a crash leaving dirty files (broken syntax).
        fs::write(wt_path.join("src/lib.rs"), "THIS IS BROKEN SYNTAX!!!\n").unwrap();
        // Also add an untracked file.
        fs::write(wt_path.join("crash_residue.txt"), "untracked junk").unwrap();

        // cargo_check_in_worktree should stash dirty files, check committed code, and restore.
        assert!(
            cargo_check_in_worktree(&wt_path),
            "cargo check should pass on committed code after stashing dirty files"
        );

        // Bug 651: dirty files must be restored after cargo check.
        assert_eq!(
            fs::read_to_string(wt_path.join("src/lib.rs")).unwrap(),
            "THIS IS BROKEN SYNTAX!!!\n",
            "modified tracked file should be restored after cargo check"
        );
        assert!(
            wt_path.join("crash_residue.txt").exists(),
            "untracked file should be restored after cargo check"
        );
    }

    // ── GateFailureKind::classify ─────────────────────────────────────────────

    #[test]
    fn classify_fmt_from_diff_in() {
        assert_eq!(
            GateFailureKind::classify("Diff in server/src/lib.rs\n--- original\n+++ reformatted"),
            GateFailureKind::Fmt
        );
    }

    #[test]
    fn classify_fmt_from_would_reformat() {
        assert_eq!(
            GateFailureKind::classify(
                "Checking server/src/lib.rs\nwould reformat server/src/lib.rs"
            ),
            GateFailureKind::Fmt
        );
    }

    #[test]
    fn classify_lint_from_clippy_error() {
        assert_eq!(
            GateFailureKind::classify("error[clippy::unused_variable]: unused variable `x`"),
            GateFailureKind::Lint
        );
    }

    #[test]
    fn classify_lint_from_clippy_warning() {
        assert_eq!(
            GateFailureKind::classify("warning[clippy::needless_return]: unneeded return"),
            GateFailureKind::Lint
        );
    }

    #[test]
    fn classify_lint_from_missing_doc_comments() {
        assert_eq!(
            GateFailureKind::classify(
                "error: missing_doc_comments: public item lacks documentation"
            ),
            GateFailureKind::Lint
        );
    }

    /// Story 1101: a passing test whose name contains a lint trigger keyword
    /// must NOT produce a Lint classification.
    #[test]
    fn classify_does_not_false_positive_on_test_name_substring() {
        let output = "test agents::gates::tests::classify_lint_from_missing_doc_comments ... ok\n\
             test result: ok. 1 passed; 0 failed";
        assert_ne!(
            GateFailureKind::classify(output),
            GateFailureKind::Lint,
            "passing test name containing 'missing_doc_comments' must not classify as Lint"
        );
    }

    #[test]
    fn classify_source_map_check_from_missing_docs_direction() {
        assert_eq!(
            GateFailureKind::classify("missing-docs direction: server/src/foo.rs:42 pub fn bar"),
            GateFailureKind::SourceMapCheck
        );
    }

    #[test]
    fn classify_content_conflict() {
        assert_eq!(
            GateFailureKind::classify("CONFLICT (content): Merge conflict in server/src/lib.rs"),
            GateFailureKind::ContentConflict
        );
    }

    #[test]
    fn classify_build_from_rustc_compile_error() {
        // Post-squash compile errors (typical when master drifts under a feature
        // branch — e.g. story 1018 hit `error[E0063]: missing field` after
        // master gained a Stage::Coding field the feature branch did not set).
        assert_eq!(
            GateFailureKind::classify(
                "error[E0063]: missing field `plan` in initializer of `Stage`\n   --> server/src/foo.rs:166:20"
            ),
            GateFailureKind::Build
        );
    }

    #[test]
    fn classify_build_does_not_misfire_on_clippy_error() {
        // Clippy errors look like `error[clippy::name]` and must remain Lint,
        // not Build, because the `error[E` prefix would otherwise overlap.
        assert_eq!(
            GateFailureKind::classify("error[clippy::unused_variable]: unused variable `x`"),
            GateFailureKind::Lint
        );
    }

    #[test]
    fn classify_test_failure_for_unrecognised_output() {
        assert_eq!(
            GateFailureKind::classify("test result: FAILED. 3 passed; 1 failed"),
            GateFailureKind::Test
        );
    }

    // ── GateFailureKind::is_self_evident_fix ─────────────────────────────────

    #[test]
    fn fmt_is_self_evident_fix() {
        assert!(GateFailureKind::Fmt.is_self_evident_fix());
    }

    #[test]
    fn lint_is_self_evident_fix() {
        assert!(GateFailureKind::Lint.is_self_evident_fix());
    }

    #[test]
    fn source_map_check_is_self_evident_fix() {
        assert!(GateFailureKind::SourceMapCheck.is_self_evident_fix());
    }

    #[test]
    fn test_failure_is_not_self_evident_fix() {
        assert!(!GateFailureKind::Test.is_self_evident_fix());
    }

    #[test]
    fn content_conflict_is_not_self_evident_fix() {
        assert!(!GateFailureKind::ContentConflict.is_self_evident_fix());
    }

    #[test]
    fn build_error_is_not_self_evident_fix() {
        assert!(!GateFailureKind::Build.is_self_evident_fix());
    }

    #[test]
    fn other_is_not_self_evident_fix() {
        assert!(!GateFailureKind::Other.is_self_evident_fix());
    }

    // ── GateOutcome constructors ──────────────────────────────────────────────

    #[test]
    fn gate_outcome_pass_has_no_failure_kind() {
        let outcome = GateOutcome::pass("all tests passed".to_string());
        assert!(outcome.passed);
        assert!(outcome.failure_kind.is_none());
        assert_eq!(outcome.output, "all tests passed");
    }

    #[test]
    fn gate_outcome_fail_classifies_kind() {
        let outcome = GateOutcome::fail("Diff in server/src/lib.rs".to_string());
        assert!(!outcome.passed);
        assert_eq!(outcome.failure_kind, Some(GateFailureKind::Fmt));
    }

    #[test]
    fn gate_outcome_build_error_sets_build_kind() {
        let outcome = GateOutcome::build_error("ERROR [E0761]: duplicate module files".to_string());
        assert!(!outcome.passed);
        assert_eq!(outcome.failure_kind, Some(GateFailureKind::Build));
    }
}