linersh 1.1.2 → 1.2.0

package/dist/agents/runner.js

@@ -1,906 +0,0 @@
-import { spawn } from "node:child_process";
-import { existsSync, mkdirSync, readdirSync, readFileSync, statSync, writeFileSync } from "node:fs";
-import { join } from "node:path";
-import { validatePhaseArtifact } from "./artifact-validation.js";
-import { groupCandidateLonglist, readCandidateLonglist } from "./candidate-longlist.js";
-import { ensureEvaluationArtifact, validateEvaluationFragment } from "./evaluation-assembly.js";
-import { buildEvaluationSectionPrompt, buildPhasePrompt, WORKSPACE_DISCIPLINE } from "./prompts.js";
-import { LineBuffer, parseLine } from "./parsers.js";
-import { closeRunLog, openRunLog } from "./run-log.js";
-import { resolveRunProfile } from "./models.js";
-import { readConfig } from "../config.js";
-// How long to wait for the agent's first byte of output before declaring it
-// unreachable. Generous enough to cover a cold CLI start + auth refresh, short
-// enough that a dead subscription doesn't hang the TUI indefinitely.
-const WATCHDOG_MS = 15_000;
-const DEFAULT_EVALUATION_CLOSURE_IDLE_MS = 120_000;
-/**
- * Spawn the configured agent CLI with a phase-specific prompt. The agent's
- * newline-delimited JSON stdout is parsed into AgentEvents so the PhaseRunner
- * can render tool calls / streaming text / summary as structured panes.
- *
- * Thin wrapper over `runAgentTask` — builds the per-phase prompt via
- * `buildPhasePrompt` and labels the audit log with the phase id.
- */
-export function runPhaseWithAgent(args) {
-    const recoverableArtifacts = new Set([
-        "candidates",
-        "evaluation",
-        "quality",
-        "synthesis",
-        "assembly",
-    ]);
-    if (args.resume && recoverableArtifacts.has(args.phaseId)) {
-        const existing = ensurePhaseArtifact(args.project, args.phaseId);
-        if (existing.ok) {
-            const label = args.phaseId.charAt(0).toUpperCase() + args.phaseId.slice(1);
-            args.onEvent({
-                kind: "raw",
-                text: `[liner] ${label} artifact already valid; continuing without another AI run.`,
-            });
-            return { cancel: () => { }, done: Promise.resolve({ code: 0, stderr: "" }) };
-        }
-    }
-    if (args.phaseId === "evaluation" && (!args.resume || hasSectionedEvaluationState(args.project))) {
-        return runEvaluationSectionsWithAgent(args);
-    }
-    const prompt = buildPhasePrompt({
-        phaseId: args.phaseId,
-        project: args.project,
-        skillPath: args.skillPath,
-        tape: args.tape,
-    });
-    // Per-phase model: the built-in map downgrades the heavy phases, with the
-    // user's config.yaml overrides winning. The compile phase isn't an agent
-    // run; gates don't reach here either.
-    const config = readConfig();
-    const overrides = config.models?.[args.agent.id];
-    const profile = resolveRunProfile(args.agent.id, args.phaseId, overrides, config.providerPreferences?.[args.agent.id]);
-    const handle = runAgentTask({
-        agent: args.agent,
-        project: args.project,
-        skillPath: args.skillPath,
-        prompt,
-        model: profile.model,
-        reasoningEffort: profile.reasoningEffort,
-        modelFallbackSource: profile.modelSource === "builtin" || profile.modelSource === "auto" ? profile.modelSource : undefined,
-        allowEffortFallback: profile.effortSource === "auto",
-        taskLabel: args.phaseId,
-        resume: args.resume,
-        onEvent: args.onEvent,
-    });
-    return {
-        cancel: handle.cancel,
-        done: handle.done.then((result) => {
-            if (result.code !== 0)
-                return result;
-            const validation = ensurePhaseArtifact(args.project, args.phaseId);
-            if (validation.ok) {
-                if ("message" in validation && validation.message) {
-                    args.onEvent({ kind: "raw", text: `[liner] ${validation.message}` });
-                }
-                return result;
-            }
-            const message = `[liner] Artifact validation failed after ${args.phaseId}: ${validation.message}`;
-            args.onEvent({ kind: "text", text: message });
-            return {
-                ...result,
-                code: 1,
-                stderr: result.stderr ? `${result.stderr}\n${message}` : message,
-            };
-        }),
-    };
-}
-function runEvaluationSectionsWithAgent(args) {
-    if (args.resume) {
-        const existingEvaluation = ensureEvaluationArtifact(args.project);
-        if (existingEvaluation.ok) {
-            const message = existingEvaluation.message || "Evaluation artifact already complete.";
-            args.onEvent({ kind: "raw", text: `[liner] ${message}` });
-            return { cancel: () => { }, done: Promise.resolve({ code: 0, stderr: "" }) };
-        }
-    }
-    const groups = groupCandidateLonglist(readCandidateLonglist(join(args.project, "working/02-candidate-longlist.md")));
-    if (groups.length === 0) {
-        const message = "[liner] Cannot run Evaluation: working/02-candidate-longlist.md has no URL candidates.";
-        args.onEvent({ kind: "raw", text: message });
-        return { cancel: () => { }, done: Promise.resolve({ code: 1, stderr: message }) };
-    }
-    const config = readConfig();
-    const overrides = config.models?.[args.agent.id];
-    const profile = resolveRunProfile(args.agent.id, "evaluation", overrides, config.providerPreferences?.[args.agent.id]);
-    let cancelCurrent = () => { };
-    let cancelled = false;
-    let lastLogPath;
-    const done = (async () => {
-        markSectionedEvaluationStarted(args.project, groups.length);
-        args.onEvent({
-            kind: "raw",
-            text: `[liner] Running Evaluation in ${groups.length} section chunk${groups.length === 1 ? "" : "s"}.`,
-        });
-        for (const group of groups) {
-            if (cancelled)
-                return { code: 1, stderr: "[liner] Evaluation cancelled." };
-            const fragment = validateEvaluationFragment(join(args.project, group.fragmentPath), group.candidates);
-            if (fragment.ok) {
-                args.onEvent({
-                    kind: "raw",
-                    text: `[liner] Evaluation chunk ${group.index}/${group.total} already complete: ` +
-                        `${group.section} (${fragment.count} candidates).`,
-                });
-                continue;
-            }
-            args.onEvent({
-                kind: "raw",
-                text: `[liner] Evaluation chunk ${group.index}/${group.total}: ` +
-                    `${group.section} (${group.candidates.length} candidates).`,
-            });
-            const handle = runAgentTask({
-                agent: args.agent,
-                project: args.project,
-                skillPath: args.skillPath,
-                prompt: buildEvaluationSectionPrompt({
-                    project: args.project,
-                    skillPath: args.skillPath,
-                    tape: args.tape,
-                    group,
-                }),
-                model: profile.model,
-                reasoningEffort: profile.reasoningEffort,
-                modelFallbackSource: profile.modelSource === "builtin" || profile.modelSource === "auto" ? profile.modelSource : undefined,
-                allowEffortFallback: profile.effortSource === "auto",
-                taskLabel: "evaluation",
-                onEvent: args.onEvent,
-            });
-            cancelCurrent = handle.cancel;
-            const result = await handle.done;
-            lastLogPath = result.logPath;
-            if (result.code !== 0)
-                return result;
-        }
-        const validation = ensureEvaluationArtifact(args.project);
-        if (!validation.ok) {
-            const message = `[liner] Artifact validation failed after evaluation: ${validation.message}`;
-            args.onEvent({ kind: "text", text: message });
-            return { code: 1, stderr: message, logPath: lastLogPath };
-        }
-        if (validation.message) {
-            args.onEvent({ kind: "raw", text: `[liner] ${validation.message}` });
-        }
-        return { code: 0, stderr: "", logPath: lastLogPath };
-    })();
-    return {
-        cancel: () => {
-            cancelled = true;
-            cancelCurrent();
-        },
-        done,
-    };
-}
-function hasSectionedEvaluationState(project) {
-    if (existsSync(join(project, "working/evaluation-decisions/.liner-sectioned-evaluation.json"))) {
-        return true;
-    }
-    const dir = join(project, "working/evaluation-decisions");
-    if (!existsSync(dir))
-        return false;
-    return readdirSync(dir).some((file) => /\.ya?ml$/i.test(file));
-}
-function markSectionedEvaluationStarted(project, chunkCount) {
-    const dir = join(project, "working/evaluation-decisions");
-    mkdirSync(dir, { recursive: true });
-    writeFileSync(join(dir, ".liner-sectioned-evaluation.json"), JSON.stringify({
-        version: 1,
-        chunkCount,
-        updatedAt: new Date().toISOString(),
-    }, null, 2) + "\n", "utf8");
-}
-function ensurePhaseArtifact(project, phaseId) {
-    if (phaseId !== "evaluation")
-        return validatePhaseArtifact(project, phaseId);
-    return ensureEvaluationArtifact(project);
-}
-/**
- * Run a single agent task with a caller-supplied prompt. Used by methodology
- * phases (via `runPhaseWithAgent`) and by one-off side tasks like the JTBD
- * elicitation step in the new-mixtape wizard. The agent's stdout is parsed
- * the same way (stream JSON → AgentEvent) so callers get tool calls + final
- * text via the same callback interface.
- */
-export function runAgentTask(args) {
-    // Delegating cancel: a model-rejection fallback re-spawns under the hood, so
-    // the handle's cancel must target whichever child is currently live.
-    let cancelCurrent = () => { };
-    const done = runAttempt(args, {}, (c) => {
-        cancelCurrent = c;
-    });
-    return { cancel: () => cancelCurrent(), done };
-}
-const CLAUDE_METHOD_TOOLS = "Read Write Edit Glob Grep WebFetch";
-const CLAUDE_METHOD_BUILTIN_TOOLS = "Read,Write,Edit,Glob,Grep,WebFetch";
-const CLAUDE_HARD_DENIED_TOOLS = "Bash Task WebSearch ToolSearch";
-function isolatedImprovementDeniedTools(workspace) {
-    const corpus = portableDirname(portableDirname(portableDirname(workspace)));
-    return [
-        CLAUDE_HARD_DENIED_TOOLS,
-        `Edit(${claudeAbsolutePermissionPath(portableJoin(corpus, "tape.yaml"))})`,
-        `Edit(${claudeAbsolutePermissionPath(portableJoin(corpus, "synthesis.md"))})`,
-        `Edit(${claudeAbsolutePermissionPath(portableJoin(corpus, "working"))}/**)`,
-    ].join(" ");
-}
-function portableDirname(path) {
-    const normalized = path.replace(/\\/g, "/").replace(/\/+$/, "");
-    const index = normalized.lastIndexOf("/");
-    if (index < 0)
-        return ".";
-    if (index === 0)
-        return "/";
-    return normalized.slice(0, index);
-}
-function portableJoin(parent, child) {
-    return `${parent.replace(/\\/g, "/").replace(/\/+$/, "")}/${child}`;
-}
-function claudeAbsolutePermissionPath(path) {
-    const normalized = path.replace(/\\/g, "/");
-    if (/^[A-Za-z]:\//.test(normalized))
-        return `//${normalized}`;
-    if (normalized.startsWith("/"))
-        return `/${normalized}`;
-    return `//${normalized}`;
-}
-/**
- * One spawn attempt. Built-in choices and the Auto model policy may retry once on the provider
- * default after an unsupported model or effort. User-selected options fail
- * clearly without substitution. `registerCancel` is re-invoked per attempt so
- * the caller's cancel always hits the live child.
- */
-function runAttempt(args, attempt, registerCancel) {
-    const command = args.agent.bin;
-    const resume = args.resume === true;
-    const resumeSessionId = args.agent.id === "codex" && resume
-        ? findLatestCodexSessionId(args.project, args.taskLabel)
-        : undefined;
-    if (args.agent.id === "codex" && resume && !resumeSessionId) {
-        const message = `[liner] Cannot resume OpenAI ${args.taskLabel}: no prior Codex CLI session id was found ` +
-            `in .liner-runs/${args.taskLabel}. Retry this phase from a fresh run instead.`;
-        args.onEvent({ kind: "raw", text: message });
-        const log = openRunLog(args.project, args.taskLabel, {
-            agent: args.agent.id,
-            resume,
-        });
-        log.write(JSON.stringify({ type: "_liner_error", message }));
-        closeRunLog(log, 1, message.length);
-        return Promise.resolve({ code: 1, stderr: message, logPath: log.path });
-    }
-    const cliArgs = buildArgs(args.agent.id, args.project, args.skillPath, resume, args.model, resumeSessionId, args.taskLabel, args.reasoningEffort);
-    const child = spawn(command, cliArgs, {
-        stdio: ["pipe", "pipe", "pipe"],
-        cwd: args.project,
-        env: {
-            ...process.env,
-            ...(args.agent.configHome
-                ? {
-                    [args.agent.id === "claude" ? "CLAUDE_CONFIG_DIR" : "CODEX_HOME"]: args.agent.configHome,
-                }
-                : {}),
-            // Disable extended/interleaved thinking for the spawned Claude Code.
-            // A long multi-turn tool loop with thinking on hits a hard API 400 once
-            // Claude Code auto-compacts the conversation: a prior `thinking` block
-            // gets modified on replay and the API rejects the next turn
-            // ("`thinking` or `redacted_thinking` blocks ... cannot be modified").
-            // It surfaces both on fresh runs (~25+ turns) and instantly on resume
-            // (the resume replays the already-corrupted message). Turning thinking
-            // off removes the blocks entirely, so the failure class can't occur.
-            // Only Claude reads this; Codex ignores it.
-            ...(args.agent.id === "claude" ? { MAX_THINKING_TOKENS: "0" } : {}),
-        },
-    });
-    let cancelled = false;
-    registerCancel(() => {
-        cancelled = true;
-        if (!child.killed)
-            child.kill("SIGTERM");
-    });
-    // Accumulated tail of the agent's own output (stdout + stderr). Used only to
-    // recognize a model-rejection error for the fallback; capped so a long run
-    // doesn't retain the whole transcript.
-    let outputTail = "";
-    const noteForDetection = (chunk) => {
-        outputTail = (outputTail + chunk).slice(-4000);
-    };
-    // On resume, the previous session already contains the prompt + all prior
-    // tool-call results in context. Send a tiny continuation nudge instead of
-    // re-shipping the full prompt — that's the whole point of resume (cheaper,
-    // no re-fetching).
-    const stdinPayload = resume
-        ? args.resumePrompt ?? resumePromptForTask(args.taskLabel)
-        : args.prompt;
-    child.stdin?.write(stdinPayload, "utf8");
-    child.stdin?.end();
-    // Tee-style audit log of the raw JSONL stream. Not load-bearing (Claude's
-    // session.json already covers resume) — this gives us a per-run transcript
-    // the curator can scroll back through and a forensic record when things
-    // misbehave.
-    const log = openRunLog(args.project, args.taskLabel, {
-        agent: args.agent.id,
-        resume,
-    });
-    // Agent-unavailability watchdog. A dead subscription / revoked auth / model
-    // deprecation can leave the agent process launched but mute — no stream
-    // events, no stderr, no exit — and the TUI would otherwise spin forever. If
-    // nothing comes back within the window, surface a clear error and kill the
-    // process so the run lands in the failed state (where retry / switch-agent
-    // live). Any output at all — stdout or stderr — clears the watchdog.
-    let sawOutput = false;
-    const watchdog = setTimeout(() => {
-        if (sawOutput || child.killed)
-            return;
-        const name = args.agent.id === "claude" ? "Claude" : "OpenAI";
-        args.onEvent({
-            kind: "raw",
-            text: `[liner] No response from ${name} after ${WATCHDOG_MS / 1000}s. ` +
-                `It may not be signed in, or its subscription/quota may be exhausted. ` +
-                `Try running \`${args.agent.id}\` on its own to check, then retry or switch agents.`,
-        });
-        log.write(JSON.stringify({ type: "_liner_watchdog_timeout", ms: WATCHDOG_MS }));
-        child.kill("SIGTERM");
-    }, WATCHDOG_MS);
-    watchdog.unref?.();
-    const noteOutput = () => {
-        if (sawOutput)
-            return;
-        sawOutput = true;
-        clearTimeout(watchdog);
-    };
-    const closureIdleMs = evaluationClosureIdleMs();
-    const canCloseEvaluationArtifact = args.taskLabel === "evaluation" &&
-        !attempt.artifactClosure &&
-        closureIdleMs > 0;
-    let evaluationClosureTimer;
-    let evaluationClosureTimedOut = false;
-    let childClosed = false;
-    let forcedEvaluationResult;
-    const activeTools = new Set();
-    const clearEvaluationClosureTimer = () => {
-        if (!evaluationClosureTimer)
-            return;
-        clearTimeout(evaluationClosureTimer);
-        evaluationClosureTimer = undefined;
-    };
-    const scheduleEvaluationClosureCheck = () => {
-        if (!canCloseEvaluationArtifact || cancelled || child.killed || childClosed)
-            return;
-        clearEvaluationClosureTimer();
-        evaluationClosureTimer = setTimeout(() => {
-            if (cancelled || child.killed)
-                return;
-            if (activeTools.size > 0) {
-                scheduleEvaluationClosureCheck();
-                return;
-            }
-            const validation = ensureEvaluationArtifact(args.project);
-            if (validation.ok) {
-                forcedEvaluationResult = {
-                    code: 0,
-                    message: "[liner] Evaluation artifact is valid; stopping the quiet agent turn.",
-                };
-                args.onEvent({ kind: "raw", text: forcedEvaluationResult.message });
-                log.write(JSON.stringify({
-                    type: "_liner_artifact_closure_complete",
-                    phase: "evaluation",
-                    ms: closureIdleMs,
-                }));
-                child.kill("SIGTERM");
-                return;
-            }
-            evaluationClosureTimedOut = true;
-            const idleLabel = closureIdleMs >= 1000 ? `${Math.round(closureIdleMs / 1000)}s` : `${closureIdleMs}ms`;
-            const message = `[liner] Evaluation has been quiet for ${idleLabel} ` +
-                `with no valid working/03-evaluation.yaml. Stopping this turn and resuming once ` +
-                `with a write-only artifact closure prompt.`;
-            args.onEvent({ kind: "raw", text: message });
-            log.write(JSON.stringify({
-                type: "_liner_artifact_closure_timeout",
-                phase: "evaluation",
-                ms: closureIdleMs,
-                validation: validation.message,
-            }));
-            child.kill("SIGTERM");
-        }, closureIdleMs);
-        evaluationClosureTimer.unref?.();
-    };
-    const noteAgentEvent = (event) => {
-        if (!canCloseEvaluationArtifact)
-            return;
-        if (event.kind === "tool_start" && event.id) {
-            activeTools.add(event.id);
-            scheduleEvaluationClosureCheck();
-            return;
-        }
-        if (event.kind === "tool_done" && event.id) {
-            activeTools.delete(event.id);
-            scheduleEvaluationClosureCheck();
-            return;
-        }
-        if (event.kind === "summary") {
-            clearEvaluationClosureTimer();
-            return;
-        }
-        scheduleEvaluationClosureCheck();
-    };
-    if (args.taskLabel === "evaluation" && attempt.artifactClosure && closureIdleMs > 0) {
-        clearEvaluationClosureTimer();
-        evaluationClosureTimer = setTimeout(() => {
-            if (cancelled || child.killed || childClosed)
-                return;
-            const validation = ensureEvaluationArtifact(args.project);
-            const idleLabel = closureIdleMs >= 1000 ? `${Math.round(closureIdleMs / 1000)}s` : `${closureIdleMs}ms`;
-            if (validation.ok) {
-                forcedEvaluationResult = {
-                    code: 0,
-                    message: `[liner] Evaluation artifact closure produced a valid YAML file within ${idleLabel}; stopping the idle closure turn.`,
-                };
-                args.onEvent({ kind: "raw", text: forcedEvaluationResult.message });
-                log.write(JSON.stringify({
-                    type: "_liner_artifact_closure_complete",
-                    phase: "evaluation",
-                    ms: closureIdleMs,
-                }));
-                child.kill("SIGTERM");
-                return;
-            }
-            forcedEvaluationResult = {
-                code: 1,
-                message: `[liner] Evaluation artifact closure timed out after ${idleLabel}: ${validation.message}. ` +
-                    "Retry Evaluation fresh or reduce the candidate set before continuing.",
-            };
-            args.onEvent({ kind: "raw", text: forcedEvaluationResult.message });
-            log.write(JSON.stringify({
-                type: "_liner_artifact_closure_failed",
-                phase: "evaluation",
-                ms: closureIdleMs,
-                validation: validation.message,
-            }));
-            child.kill("SIGTERM");
-        }, closureIdleMs);
-        evaluationClosureTimer.unref?.();
-    }
-    const buffer = new LineBuffer();
-    if (child.stdout) {
-        child.stdout.setEncoding("utf8");
-        child.stdout.on("data", (chunk) => {
-            noteOutput();
-            noteForDetection(chunk);
-            const lines = buffer.push(chunk);
-            for (const line of lines) {
-                log.write(line);
-                for (const event of parseLine(args.agent.id, line)) {
-                    noteAgentEvent(event);
-                    args.onEvent(event);
-                }
-            }
-        });
-    }
-    let stderr = "";
-    if (child.stderr) {
-        child.stderr.setEncoding("utf8");
-        child.stderr.on("data", (chunk) => {
-            noteOutput();
-            noteForDetection(chunk);
-            stderr += chunk;
-            // Surface meaningful stderr as raw events so users see install/auth
-            // hints inline, while keeping repetitive cosmetic Codex startup warnings
-            // out of the live TUI. The full stream is still recorded in the run log.
-            for (const line of chunk.split("\n")) {
-                const trimmed = line.trim();
-                if (trimmed) {
-                    // Record stderr in the log too — wrapped so it doesn't masquerade
-                    // as a real JSON event when the log is replayed.
-                    log.write(JSON.stringify({ type: "_liner_stderr", text: trimmed }));
-                    if (shouldSurfaceAgentStderr(args.agent.id, trimmed)) {
-                        args.onEvent({ kind: "raw", text: trimmed });
-                    }
-                }
-            }
-        });
-    }
-    return new Promise((resolve) => {
-        child.on("close", (code) => {
-            clearTimeout(watchdog);
-            clearEvaluationClosureTimer();
-            childClosed = true;
-            // Flush any trailing partial line.
-            for (const line of buffer.flush()) {
-                log.write(line);
-                for (const event of parseLine(args.agent.id, line)) {
-                    noteAgentEvent(event);
-                    args.onEvent(event);
-                }
-            }
-            closeRunLog(log, code, stderr.length);
-            if (forcedEvaluationResult) {
-                resolve({
-                    code: forcedEvaluationResult.code,
-                    stderr: forcedEvaluationResult.code === 0
-                        ? stderr
-                        : stderr
-                            ? `${stderr}\n${forcedEvaluationResult.message}`
-                            : forcedEvaluationResult.message,
-                    logPath: log.path,
-                });
-                return;
-            }
-            if (evaluationClosureTimedOut && !cancelled) {
-                args.onEvent({
-                    kind: "raw",
-                    text: "[liner] Resuming Evaluation once to close working/03-evaluation.yaml. No more source fetching in this pass.",
-                });
-                resolve(runAttempt({
-                    ...args,
-                    resume: true,
-                    resumePrompt: evaluationArtifactClosurePrompt(),
-                }, { artifactClosure: true }, registerCancel));
-                return;
-            }
-            // Configured-option rejection. Auto-owned options may retry once on the
-            // provider default; user-owned options fail closed. Resumes and
-            // cancellations are exempt, and both classifiers require plausible
-            // provider rejection output so auth or quota failures surface normally.
-            if (code !== 0 &&
-                !cancelled &&
-                !resume &&
-                args.reasoningEffort &&
-                looksLikeEffortRejection(outputTail, args.reasoningEffort)) {
-                if (!attempt.modelFallback && args.allowEffortFallback === true) {
-                    const policy = args.modelFallbackSource === "auto" ? "the Auto model policy" : "Liner's default Thinking effort";
-                    args.onEvent({
-                        kind: "raw",
-                        text: `[liner] ${args.agent.name} rejected ${policy} for ${args.taskLabel} — ` +
-                            "retrying once on the provider default model and Thinking effort.",
-                    });
-                    log.write(JSON.stringify({
-                        type: "_liner_auto_model_policy_fallback",
-                        model: args.model,
-                        reasoningEffort: args.reasoningEffort,
-                    }));
-                    resolve(runAttempt({
-                        ...args,
-                        // Drop what Liner chose (Auto or built-in); keep a model from
-                        // Settings or a phase override.
-                        model: args.modelFallbackSource === undefined ? args.model : undefined,
-                        reasoningEffort: undefined,
-                    }, { modelFallback: true }, registerCancel));
-                    return;
-                }
-                const message = `[liner] ${args.agent.name} rejected configured Thinking effort "${args.reasoningEffort}". ` +
-                    "Choose another effort in Settings; Liner did not substitute another effort or model.";
-                args.onEvent({ kind: "raw", text: message });
-                resolve({ code, stderr: stderr ? `${message}\n${stderr}` : message, logPath: log.path });
-                return;
-            }
-            if (code !== 0 &&
-                !cancelled &&
-                !resume &&
-                !attempt.modelFallback &&
-                args.model &&
-                args.modelFallbackSource !== undefined &&
-                looksLikeModelRejection(outputTail, args.model)) {
-                const fallbackMessage = args.modelFallbackSource === "auto"
-                    ? `[liner] ${args.agent.name} rejected the Auto model policy model "${args.model}" for ${args.taskLabel} — ` +
-                        "retrying once on the provider default. To change this choice, open Settings and select another Model."
-                    : `[liner] ${args.agent.name} rejected the built-in model "${args.model}" for ${args.taskLabel} — ` +
-                        `retrying once on the provider default. To change this phase override, set ` +
-                        `models.${args.agent.id}.${args.taskLabel} in ~/.liner/config.yaml.`;
-                args.onEvent({
-                    kind: "raw",
-                    text: fallbackMessage,
-                });
-                log.write(JSON.stringify({ type: "_liner_model_fallback", model: args.model }));
-                resolve(runAttempt({
-                    ...args,
-                    model: undefined,
-                    reasoningEffort: args.allowEffortFallback === true ? undefined : args.reasoningEffort,
-                }, { modelFallback: true }, registerCancel));
-                return;
-            }
-            if (code !== 0 &&
-                !cancelled &&
-                !resume &&
-                args.model &&
-                args.modelFallbackSource === undefined &&
-                looksLikeModelRejection(outputTail, args.model)) {
-                const message = `[liner] ${args.agent.name} rejected configured model "${args.model}". ` +
-                    "Choose another model in Settings; Liner did not substitute another model.";
-                args.onEvent({ kind: "raw", text: message });
-                resolve({ code, stderr: stderr ? `${message}\n${stderr}` : message, logPath: log.path });
-                return;
-            }
-            resolve({ code, stderr, logPath: log.path });
-        });
-        child.on("error", (e) => {
-            clearTimeout(watchdog);
-            clearEvaluationClosureTimer();
-            args.onEvent({ kind: "raw", text: `[runner error] ${e.message}` });
-            log.write(JSON.stringify({ type: "_liner_error", message: e.message }));
-            closeRunLog(log, 1, stderr.length);
-            resolve({ code: 1, stderr: e.message, logPath: log.path });
-        });
-    });
-}
-export function resumePromptForTask(taskLabel) {
-    const base = "Continue from where you left off. Pick up the same task; do not re-do tool calls you already completed." +
-        `\n\n${WORKSPACE_DISCIPLINE}`;
-    if (taskLabel === "candidates") {
-        return `${base}\n\nPhase 2 reminder: stop any open-ended web/search loop. If working/05-operating-fit-audit.md exists with status: improvement_recommended, treat this as a focused improvement pass and search for the missing source roles named there. Otherwise use the verified URLs, curator-provided tape sources, titles, and reasons already gathered in this conversation. Write working/02-candidate-longlist.md now, grouped by the knowledge-map sections, replacing the placeholder completely. Do not run more than one final targeted search for a genuinely empty section; if a section is still thin, document the best verified candidates you have and stop after writing the file.`;
-    }
-    if (taskLabel === "evaluation") {
-        return `${base}\n\nPhase 4 reminder: do not get stuck chasing unavailable content. For any candidate whose content is still unavailable, make at most one more recovery attempt, then evaluate from verified metadata, the Phase 2 reason, and any partial content. Every candidate still needs a keep/trim/drop decision. Kept/trim entries need rating, jtbd_fit (direct/bridge/background), section, rationale, fetch_status (readable/partial only), content_quality (high/medium only), at least two content-specific evidence bullets, and the three-part curator note. Do not keep sources from URL/title/search snippets/model memory alone. Prefer section-sized fragments under working/evaluation-decisions/ so Liner can assemble working/03-evaluation.yaml; if you write the final file directly, do not say you are ready to write the artifact before writing it.`;
-    }
-    if (taskLabel === "quality") {
-        return `${base}\n\nPhase 5 reminder: stop any open-ended web/search loop. Quality is a bounded audit, not fresh discovery. Use the existing kept/trim decisions in working/03-evaluation.yaml, assign missing jtbd_fit and source kinds from the title/note/section/source type, repair weak curator notes from existing evidence, spend no more external search attempts, and write working/04-quality-checks.md now with Test 0 core-action fit, Tests 1-6, Perspectives audit, the kind distribution, and the note-quality check. If a material source-role gap remains, write working/05-operating-fit-audit.md with status: improvement_recommended and concrete search lanes for a focused second pass; do not call the corpus ready with a limitation.`;
-    }
-    return base;
-}
-export function evaluationArtifactClosurePrompt() {
-    return [
-        "Continue Phase 4, but switch to artifact closure only.",
-        "",
-        WORKSPACE_DISCIPLINE,
-        "",
-        "Do not fetch, search, or read external sources in this pass. Use the evidence, metadata, Phase 2 rationale, and partial content already gathered in this conversation.",
-        "Write complete section-sized YAML fragments under working/evaluation-decisions/ now, or write working/03-evaluation.yaml directly if it is small enough.",
-        "Every URL candidate in working/02-candidate-longlist.md must appear exactly once across those fragments or in the final file.",
-        "Use decision: dropped for uncertain, duplicate, unavailable, or metadata-only candidates unless the prior evidence clearly supports kept or trim.",
-        "For kept/trim entries include rating, jtbd_fit (direct/bridge/background), fetch_status (readable/partial only), content_quality (high/medium only), at least two content-specific evidence bullets, section, rationale, and the three-part curator note. For dropped entries include URL, title, decision, section if known, and one rationale sentence.",
-        "Do not keep sources from URL/title/search snippets/model memory alone; if evidence is not already in the conversation, mark the candidate dropped in this closure pass.",
-        "After writing fragments, Liner will assemble working/03-evaluation.yaml; if you write the final file yourself, parse/check it before the required Phase 4 final message.",
-    ].join("\n");
-}
-export function evaluationClosureIdleMs(env = process.env) {
-    const raw = env["LINER_EVALUATION_CLOSURE_IDLE_MS"];
-    if (raw == null || raw.trim() === "")
-        return DEFAULT_EVALUATION_CLOSURE_IDLE_MS;
-    const parsed = Number(raw);
-    if (!Number.isFinite(parsed) || parsed < 0)
-        return DEFAULT_EVALUATION_CLOSURE_IDLE_MS;
-    return parsed;
-}
-export function buildArgs(id, project, skillPath, resume, model, resumeSessionId, taskLabel, reasoningEffort) {
-    const isolatedImprovement = taskLabel === "improvement";
-    if (id === "claude") {
-        // Stream-json requires --verbose. --include-partial-messages would give
-        // us mid-message chunking but the cost is ~10× more events; we let the
-        // assistant message arrive as one block per turn instead.
-        //
-        // Tool scoping keeps the agent's reach bounded even with permissions
-        // skipped. --tools constrains Claude's built-in tool surface shown in the
-        // init stream, while --allowedTools/--disallowedTools guard permission
-        // checks and any globally-provided names that may appear despite the
-        // built-in list. The methodology needs local file reads/writes plus
-        // targeted fetches; it should not shell out, launch subagents, or web-search.
-        //
-        // `--continue` (resume mode) picks up the most recent session in the
-        // current working directory. Since we spawn with cwd=project, "most
-        // recent session" is the one we started for this phase. The previous
-        // conversation — prompt, tool calls, partial work — stays in context.
-        //
-        // `--strict-mcp-config` with no `--mcp-config` loads zero MCP servers, so
-        // the methodology agent doesn't inherit the user's global connectors
-        // (Gmail, Drive, etc.). Those are irrelevant to curation, bloat the
-        // context window, and widen the agent's reach for no benefit.
-        const base = [
-            "-p",
-            "--output-format",
-            "stream-json",
-            "--verbose",
-            ...(!isolatedImprovement ? ["--add-dir", skillPath] : []),
-            ...(isolatedImprovement ? ["--setting-sources", "", "--permission-mode", "default"] : []),
-            "--strict-mcp-config",
-            ...(!isolatedImprovement ? ["--dangerously-skip-permissions"] : []),
-            "--tools",
-            CLAUDE_METHOD_BUILTIN_TOOLS,
-            "--allowedTools",
-            isolatedImprovement
-                ? `Read Glob Grep WebFetch Edit(${claudeAbsolutePermissionPath(project)}/**)`
-                : CLAUDE_METHOD_TOOLS,
-            "--disallowedTools",
-            isolatedImprovement ? isolatedImprovementDeniedTools(project) : CLAUDE_HARD_DENIED_TOOLS,
-            "--disable-slash-commands",
-        ];
-        // Apply the per-phase model on fresh runs only — a resumed session keeps
-        // the model it was started with, and `--model` alongside `--continue`
-        // wouldn't change the already-established conversation's model.
-        if (model && !resume)
-            base.push("--model", model);
-        if (reasoningEffort && !resume)
-            base.push("--effort", reasoningEffort);
-        if (resume)
-            base.push("--continue");
-        return base;
-    }
-    if (id === "codex") {
-        if (resume) {
-            // Codex's `exec resume` subcommand has a narrower option set than fresh
-            // exec: there is no --cd or --sandbox flag here. Use Liner's recorded
-            // session id instead of --last so a resume cannot pick up unrelated
-            // Codex work from the user's machine.
-            const sessionArg = resumeSessionId ? [resumeSessionId] : ["--last"];
-            return [
-                "exec",
-                "resume",
-                "--skip-git-repo-check",
-                "--json",
-                ...sessionArg,
-                "-",
-            ];
-        }
-        return [
-            "exec",
-            "--cd",
-            project,
-            ...(model ? ["--model", model] : []),
-            ...(reasoningEffort ? ["-c", `model_reasoning_effort=${JSON.stringify(reasoningEffort)}`] : []),
-            ...(!isolatedImprovement ? ["--add-dir", skillPath] : []),
-            "--skip-git-repo-check",
-            "-s",
-            "workspace-write",
-            "--json",
-            "-",
-        ];
-    }
-    return [];
-}
-export function findLatestCodexSessionId(project, taskLabel) {
-    const dir = join(project, ".liner-runs", taskLabel.replace(/[^a-zA-Z0-9_-]/g, "-") || "task");
-    if (!existsSync(dir))
-        return undefined;
-    const files = readdirSync(dir)
-        .filter((name) => name.endsWith(".jsonl"))
-        .map((name) => join(dir, name))
-        .filter((path) => {
-        try {
-            return statSync(path).isFile();
-        }
-        catch {
-            return false;
-        }
-    })
-        .sort((a, b) => {
-        try {
-            return statSync(b).mtimeMs - statSync(a).mtimeMs;
-        }
-        catch {
-            return 0;
-        }
-    });
-    const fallback = [];
-    for (const file of files) {
-        const candidate = codexSessionIdFromRunLog(file, false);
-        if (candidate)
-            return candidate;
-        const resumedCandidate = codexSessionIdFromRunLog(file, true);
-        if (resumedCandidate)
-            fallback.push(resumedCandidate);
-    }
-    return fallback[0];
-}
-function codexSessionIdFromRunLog(path, allowResumeLog) {
-    let metaAgent = "";
-    let metaResume = false;
-    try {
-        for (const line of readFileSync(path, "utf8").split("\n")) {
-            if (!line.trim())
-                continue;
-            let value;
-            try {
-                value = JSON.parse(line);
-            }
-            catch {
-                continue;
-            }
-            if (!value || typeof value !== "object")
-                continue;
-            const record = value;
-            if (record.type === "_liner_meta") {
-                metaAgent = typeof record.agent === "string" ? record.agent : "";
-                metaResume = record.resume === true;
-                continue;
-            }
-            if (record.type !== "thread.started")
-                continue;
-            if (metaAgent !== "codex")
-                return undefined;
-            if (metaResume && !allowResumeLog)
-                return undefined;
-            return typeof record.thread_id === "string" && record.thread_id.trim() !== ""
-                ? record.thread_id
-                : undefined;
-        }
-    }
-    catch {
-        return undefined;
-    }
-    return undefined;
-}
-/**
- * Heuristic: did this failure come from the agent not recognizing the model?
- * Matches the common "unknown/invalid model" phrasings both CLIs emit, plus
- * the offending model id echoed next to an error token. Deliberately
- * conservative — a false negative just means no auto-fallback (the normal
- * error still surfaces); we only want to claim "model rejected" when it
- * plausibly was, so genuine auth/quota failures aren't mislabeled.
- */
-export function looksLikeModelRejection(output, model) {
-    const t = output.toLowerCase();
-    if (!t)
-        return false;
-    const phrases = [
-        "unknown model",
-        "invalid model",
-        "model not found",
-        "model_not_found",
-        "no such model",
-        "unrecognized model",
-        "unsupported model",
-        "not a valid model",
-        "model does not exist",
-        "is not a supported model",
-    ];
-    if (phrases.some((p) => t.includes(p)))
-        return true;
-    // The model id echoed back alongside a generic error/invalid token.
-    const m = model.toLowerCase();
-    return (m.length > 0 &&
-        t.includes(m) &&
-        (t.includes("error") || t.includes("invalid") || t.includes("unknown") || t.includes("not found")));
-}
-export function looksLikeEffortRejection(output, _effort) {
-    const value = output.toLowerCase();
-    if (!value.trim())
-        return false;
-    const mentionsEffort = value.includes("reasoning effort") ||
-        value.includes("reasoning_effort") ||
-        value.includes("model_reasoning_effort") ||
-        value.includes("--effort") ||
-        value.includes("effort level");
-    if (!mentionsEffort)
-        return false;
-    return [
-        "invalid",
-        "unsupported",
-        "unknown",
-        "not supported",
-        "does not support",
-        "not available",
-        "unrecognized",
-    ].some((token) => value.includes(token));
-}
-export function shouldSurfaceAgentStderr(agent, line) {
-    const value = line.trim();
-    if (!value)
-        return false;
-    if (agent !== "codex")
-        return true;
-    return !isNoisyCodexStderr(value);
-}
-function isNoisyCodexStderr(line) {
-    const value = line.toLowerCase();
-    return (isCosmeticCodexStartupWarning(line) ||
-        isCodexGlobalSkillLoadWarning(value) ||
-        isNoisyCodexMcpTransport(value) ||
-        value.includes("rmcp::transport::auth") ||
-        (value.includes("oauth") && (value.includes("mcp") || value.includes("connector"))) ||
-        value.includes("write_stdin failed: stdin is closed"));
-}
-function isCodexGlobalSkillLoadWarning(value) {
-    return value.includes("failed to load skill");
-}
-function isNoisyCodexMcpTransport(value) {
-    return value.includes("rmcp::transport::worker") && value.includes("transport channel closed");
-}
-function isCosmeticCodexStartupWarning(line) {
-    return (line.includes("codex_core_plugins::manifest: ignoring interface.defaultPrompt") ||
-        line.includes("codex_core_skills::loader: ignoring interface.icon_small") ||
-        line.includes("codex_core_skills::loader: ignoring interface.icon_large"));
-}
-//# sourceMappingURL=runner.js.map