loki-mode 8.4.0 → 8.5.0

package/README.md

@@ -15,7 +15,7 @@ _The free, source-available autonomous coding agent by [Autonomi](https://www.au
 
 [Website](https://www.autonomi.dev/) | [Documentation](wiki/Home.md) | [Installation](docs/INSTALLATION.md) | [Changelog](CHANGELOG.md) | [Purple Lab -- deprecated v7.44.0](#purple-lab)
 
-**Current release: v8.4.0**
+**Current release: v8.5.0**
 
 </div>
 

package/VERSION

@@ -1 +1 @@
-8.4.0
+8.5.0

package/autonomy/lib/iteration_attribution.py

@@ -0,0 +1,181 @@
+#!/usr/bin/env python3
+"""Explain WHY a run took the iterations it took.
+
+WHY THIS EXISTS. Iterations are the direct multiplier on both wall clock and
+cost, and the Evidence Receipt reports only `{count, succeeded, failed}`. A user
+who sees "6 iterations" cannot tell real work from a gate false-positive that
+forced five redos. Neither can we, which is worse: it means we cannot tell
+whether an expensive run was the model being slow or the harness being wrong.
+
+That distinction has already bitten this project. A measured run had the agent
+claim done on EVERY iteration while a mock-integrity false positive blocked all
+six, and a start-sha bug made the council see a permanently empty diff so it
+could never vote done. Both were HARNESS defects billed to the user as model
+cost. Removing a false positive raises accuracy AND cuts iterations, which is
+the rare change that improves both axes at once -- but only if you can see it.
+
+WHAT THIS DOES NOT DO. It does not guess. Every field is derived from records
+the engine already writes, and anything not recorded is reported as unknown
+rather than inferred. An attribution that invents a reason would be worse than
+no attribution, because it would send someone optimising the wrong thing.
+
+Sources, all already written by the engine:
+  .loki/metrics/efficiency/iteration-N.json   status, duration_ms, cost_usd, model
+  .loki/events.jsonl                          iteration_complete events
+
+Usage:
+    python3 autonomy/lib/iteration_attribution.py [--loki-dir .loki] [--json]
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import os
+import sys
+
+
+def _read_json(path, default=None):
+    try:
+        with open(path, encoding="utf-8") as fh:
+            return json.load(fh)
+    except Exception:
+        return default
+
+
+def _iteration_records(loki_dir):
+    """Return per-iteration records sorted by iteration number.
+
+    Reads only what the engine already wrote. A malformed or partial record is
+    SKIPPED rather than defaulted, because a zero-cost placeholder would silently
+    understate the total and make the attribution wrong in the direction that
+    flatters us.
+    """
+    eff_dir = os.path.join(loki_dir, "metrics", "efficiency")
+    out = []
+    try:
+        names = sorted(os.listdir(eff_dir))
+    except OSError:
+        return out
+    for name in names:
+        if not (name.startswith("iteration-") and name.endswith(".json")):
+            continue
+        rec = _read_json(os.path.join(eff_dir, name))
+        if not isinstance(rec, dict) or "iteration" not in rec:
+            continue
+        out.append(rec)
+    out.sort(key=lambda r: r.get("iteration", 0))
+    return out
+
+
+def attribute(loki_dir):
+    """Split iteration cost and time into PROGRESS versus REWORK.
+
+    The definition is deliberately conservative and stated plainly, because a
+    generous definition of "progress" is how a tool flatters itself:
+
+      progress  an iteration that COMPLETED (exit 0)
+      rework    an iteration that FAILED and therefore had to be repeated
+      unknown   an iteration whose status was never recorded
+
+    This is a floor on rework, not a ceiling. An iteration that "completed" but
+    was forced to run again by a false-positive gate is counted as progress here,
+    because the engine does not currently record the blocking gate per iteration.
+    Saying so is the point: the number is honest about what it cannot see.
+    """
+    recs = _iteration_records(loki_dir)
+
+    summary = {
+        "iterations": len(recs),
+        "progress": {"count": 0, "cost_usd": 0.0, "duration_ms": 0},
+        "rework": {"count": 0, "cost_usd": 0.0, "duration_ms": 0},
+        "unknown": {"count": 0, "cost_usd": 0.0, "duration_ms": 0},
+        "cost_recorded": False,
+        "notes": [],
+    }
+
+    for r in recs:
+        status = str(r.get("status", "")).strip().lower()
+        if status == "completed":
+            bucket = "progress"
+        elif status == "failed":
+            bucket = "rework"
+        else:
+            bucket = "unknown"
+        summary[bucket]["count"] += 1
+        cost = r.get("cost_usd")
+        if isinstance(cost, (int, float)):
+            summary[bucket]["cost_usd"] += float(cost)
+            if cost > 0:
+                summary["cost_recorded"] = True
+        dur = r.get("duration_ms")
+        if isinstance(dur, (int, float)) and dur >= 0:
+            summary[bucket]["duration_ms"] += int(dur)
+
+    for b in ("progress", "rework", "unknown"):
+        summary[b]["cost_usd"] = round(summary[b]["cost_usd"], 4)
+
+    total_cost = sum(summary[b]["cost_usd"] for b in ("progress", "rework", "unknown"))
+    summary["total_cost_usd"] = round(total_cost, 4) if summary["cost_recorded"] else None
+
+    if not recs:
+        summary["notes"].append(
+            "no efficiency records: this run predates the recorder, or no iteration completed"
+        )
+    if not summary["cost_recorded"]:
+        # Distinguishing "not recorded" from a genuine $0.00 is the same honesty
+        # property the Evidence Receipt depends on. A skeptic reading $0.00
+        # concludes the artifact is fake.
+        summary["notes"].append(
+            "cost not recorded for any iteration (reported as null, not as $0.00)"
+        )
+    if summary["rework"]["count"] and summary["cost_recorded"]:
+        share = summary["rework"]["cost_usd"] / total_cost if total_cost else 0.0
+        summary["rework_cost_share"] = round(share, 4)
+        summary["notes"].append(
+            f"{summary['rework']['count']} of {len(recs)} iterations failed and were "
+            f"repeated, costing {share:.0%} of the run"
+        )
+    summary["notes"].append(
+        "rework is a FLOOR: an iteration that completed but was forced to repeat by a "
+        "gate is counted as progress, because the blocking gate is not recorded per "
+        "iteration"
+    )
+    return summary
+
+
+def _render(s):
+    lines = ["Iteration attribution", "====================", ""]
+    lines.append(f"Iterations:   {s['iterations']}")
+    lines.append(f"  progress:   {s['progress']['count']}")
+    lines.append(f"  rework:     {s['rework']['count']}")
+    if s["unknown"]["count"]:
+        lines.append(f"  unknown:    {s['unknown']['count']}")
+    if s["total_cost_usd"] is None:
+        lines.append("Cost:         not recorded")
+    else:
+        lines.append(f"Cost:         ${s['total_cost_usd']}")
+        lines.append(f"  progress:   ${s['progress']['cost_usd']}")
+        lines.append(f"  rework:     ${s['rework']['cost_usd']}")
+    lines.append("")
+    for n in s["notes"]:
+        lines.append(f"note: {n}")
+    return "\n".join(lines)
+
+
+def main():
+    ap = argparse.ArgumentParser(description=__doc__)
+    ap.add_argument("--loki-dir", default=".loki")
+    ap.add_argument("--json", action="store_true")
+    args = ap.parse_args()
+
+    s = attribute(args.loki_dir)
+    if args.json:
+        print(json.dumps(s, indent=2))
+    else:
+        print(_render(s))
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

package/autonomy/lib/proof-generator.py

@@ -1081,6 +1081,37 @@ def _build_proof(args, loki_dir, target_dir, repo_root):
 
     files_changed, diffs, diff_base_sha = _git_diffstat(target_dir, args.include_diffs)
     iterations = _collect_iterations(loki_dir)
+    # Attribute iterations to PROGRESS vs REWORK. The receipt previously reported
+    # only {count, succeeded, failed}, so a user seeing "6 iterations" could not
+    # tell real work from a gate false-positive that forced five redos -- and
+    # neither could we, which is worse, because it hides whether an expensive run
+    # was a slow model or our own harness being wrong. Measured here: an agent
+    # once claimed done on EVERY iteration while a mock-integrity false positive
+    # blocked all six.
+    #
+    # Deterministic and derived only from records the engine already writes, so
+    # this belongs with the FACTS, not the AI assessments. Failure to import or
+    # attribute leaves iterations untouched rather than guessing: a fabricated
+    # attribution would send someone optimising the wrong thing.
+    try:
+        from iteration_attribution import attribute as _attribute_iterations
+        _attr = _attribute_iterations(loki_dir)
+        if _attr.get("iterations"):
+            iterations["attribution"] = {
+                "progress": _attr["progress"],
+                "rework": _attr["rework"],
+                "rework_cost_share": _attr.get("rework_cost_share"),
+                # Stated in the artifact, not just the tool: rework is a FLOOR.
+                # An iteration that completed but was forced to repeat by a gate
+                # counts as progress, because the blocking gate is not recorded
+                # per iteration. Overstating certainty here would be worse than
+                # omitting the split.
+                "basis": "rework counts FAILED iterations only; a completed "
+                         "iteration forced to repeat by a gate is counted as "
+                         "progress, so rework is a floor",
+            }
+    except Exception:
+        pass
     spec = _collect_spec(loki_dir, target_dir)
     council = _collect_council(loki_dir)
     quality_gates = _collect_quality_gates(loki_dir)

package/autonomy/lib/silence_report.py

@@ -0,0 +1,163 @@
+#!/usr/bin/env python3
+"""Find the longest stretches where a build told the user nothing.
+
+WHY THIS EXISTS. The loudest complaints about agents in this category are not
+"it was wrong", they are "it went quiet and I could not tell if it was working":
+opencode#11112 "always stuck at Preparing write" (76 comments), continue#7143
+"not making changes to code", continue#5696 "agent does not execute functions".
+In each case the agent may well be working. The user cannot tell, so they kill
+it or leave.
+
+Wall-clock total is the wrong metric for that experience. A ten-minute build
+that reports something every twenty seconds feels fast; a four-minute build that
+goes silent for three of them feels broken. What matters is the LONGEST GAP.
+
+This reads the events the engine already writes and reports the worst silences,
+so the number can be driven down instead of guessed at.
+
+HOW TO MEASURE IT HONESTLY, learned the hard way here:
+  - Read wall clock from the engine's own event stream, never from `ps etime` on
+    a supervising process. A hung harness once reported 39 minutes for a 20
+    minute run.
+  - Separate IDLE time between sessions from SILENCE during a build. A gap
+    between two `session_resume` events is a human having lunch, not a defect,
+    and counting it makes the report useless.
+  - Report the event on each side of a gap. "Silent for 214s" is not actionable;
+    "silent for 214s after code_review_start" names the step to fix.
+
+Usage:
+    python3 autonomy/lib/silence_report.py [--events .loki/events.jsonl]
+                                           [--threshold 30] [--json]
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import sys
+from datetime import datetime
+
+# Events that mark a session boundary rather than build progress. A gap
+# alongside one of these is a human being away, not the engine being silent.
+_BOUNDARY = {
+    "session_resume",
+    "session_start",
+    "claude_hook_sessionstart",
+    "cli_command_deprecated",
+}
+
+
+def _parse(path):
+    """Yield (datetime, event_type). Malformed lines are skipped, not defaulted.
+
+    A line we cannot parse must not become a synthetic timestamp: that would
+    invent a gap or hide one, and either way the report stops being evidence.
+    """
+    out = []
+    try:
+        fh = open(path, encoding="utf-8", errors="replace")
+    except OSError:
+        return out
+    with fh:
+        for line in fh:
+            line = line.strip()
+            if not line:
+                continue
+            try:
+                d = json.loads(line)
+            except Exception:
+                continue
+            t = d.get("timestamp") or d.get("ts") or d.get("at")
+            if not isinstance(t, str) or not t:
+                continue
+            try:
+                dt = datetime.fromisoformat(t.replace("Z", "+00:00"))
+            except Exception:
+                continue
+            kind = d.get("type") or d.get("event") or "unknown"
+            out.append((dt, str(kind)))
+    out.sort(key=lambda r: r[0])
+    return out
+
+
+def analyse(events_path, threshold_s=30.0, session_gap_s=1800.0):
+    evs = _parse(events_path)
+    if len(evs) < 2:
+        return {
+            "events": len(evs),
+            "gaps_over_threshold": 0,
+            "worst": [],
+            "threshold_s": threshold_s,
+            "notes": ["fewer than two timestamped events: nothing to measure"],
+        }
+
+    gaps = []
+    skipped_boundary = 0
+    for i in range(1, len(evs)):
+        prev_t, prev_k = evs[i - 1]
+        cur_t, cur_k = evs[i]
+        secs = (cur_t - prev_t).total_seconds()
+        if secs <= 0:
+            continue
+        # A gap longer than the session window, or one touching a session
+        # boundary event, is idle time rather than in-build silence.
+        if secs >= session_gap_s or prev_k in _BOUNDARY or cur_k in _BOUNDARY:
+            skipped_boundary += 1
+            continue
+        gaps.append({"seconds": round(secs, 1), "after": prev_k, "before": cur_k})
+
+    gaps.sort(key=lambda g: g["seconds"], reverse=True)
+    over = [g for g in gaps if g["seconds"] > threshold_s]
+
+    notes = []
+    if not gaps:
+        notes.append(
+            "no in-build gaps found: this event stream is session/CLI telemetry, "
+            "not a build. Point --events at a real build's .loki/events.jsonl"
+        )
+    if skipped_boundary:
+        notes.append(
+            f"{skipped_boundary} gap(s) ignored as idle time between sessions, "
+            "not in-build silence"
+        )
+    return {
+        "events": len(evs),
+        "in_build_gaps": len(gaps),
+        "gaps_over_threshold": len(over),
+        "threshold_s": threshold_s,
+        "worst": gaps[:10],
+        "notes": notes,
+    }
+
+
+def _render(r):
+    lines = ["Silence report", "==============", ""]
+    lines.append(f"Events:            {r['events']}")
+    lines.append(f"In-build gaps:     {r.get('in_build_gaps', 0)}")
+    lines.append(f"Over {int(r['threshold_s'])}s:          {r['gaps_over_threshold']}")
+    if r["worst"]:
+        lines.append("")
+        lines.append("Longest silences:")
+        for g in r["worst"][:5]:
+            lines.append(
+                f"  {g['seconds']:>8.1f}s  after '{g['after']}' before '{g['before']}'"
+            )
+    for n in r.get("notes", []):
+        lines.append("")
+        lines.append(f"note: {n}")
+    return "\n".join(lines)
+
+
+def main():
+    ap = argparse.ArgumentParser(description=__doc__)
+    ap.add_argument("--events", default=".loki/events.jsonl")
+    ap.add_argument("--threshold", type=float, default=30.0)
+    ap.add_argument("--json", action="store_true")
+    args = ap.parse_args()
+    r = analyse(args.events, args.threshold)
+    print(json.dumps(r, indent=2) if args.json else _render(r))
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

package/dashboard/__init__.py

@@ -7,7 +7,7 @@ Modules:
     control: Session control API (start/stop/pause/resume)
 """
 
-__version__ = "8.4.0"
+__version__ = "8.5.0"
 
 # Expose the control app for easy import
 try:

package/loki-ts/dist/loki.js

@@ -1,5 +1,5 @@
 // @bun
-var S_=Object.create;var{getPrototypeOf:y_,defineProperty:oK,getOwnPropertyNames:b_}=Object;var __=Object.prototype.hasOwnProperty;function f_(Z){return this[Z]}var h_,v_,g_=(Z,X,Q)=>{var Y=Z!=null&&typeof Z==="object";if(Y){var J=X?h_??=new WeakMap:v_??=new WeakMap,z=J.get(Z);if(z)return z}Q=Z!=null?S_(y_(Z)):{};let K=X||!Z||!Z.__esModule?oK(Q,"default",{value:Z,enumerable:!0}):Q;for(let $ of b_(Z))if(!__.call(K,$))oK(K,$,{get:f_.bind(Z,$),enumerable:!0});if(Y)J.set(Z,K);return K};var VQ=(Z,X)=>()=>(X||Z((X={exports:{}}).exports,X),X.exports);var m_=(Z)=>Z;function u_(Z,X){this[Z]=m_.bind(null,X)}var c0=(Z,X)=>{for(var Q in X)oK(Z,Q,{get:X[Q],enumerable:!0,configurable:!0,set:u_.bind(X,Q)})};var p=(Z,X)=>()=>(Z&&(X=Z(Z=0)),X);var t0=import.meta.require;var IO={};c0(IO,{lokiDir:()=>j0,homeLokiDir:()=>I4,findRepoRootForVersion:()=>tK,REPO_ROOT:()=>Q8});import{resolve as s7,dirname as rK}from"path";import{fileURLToPath as p_}from"url";import{existsSync as qQ}from"fs";import{homedir as d_}from"os";function c_(){let Z=EO;for(let X=0;X<6;X++){if(qQ(s7(Z,"VERSION"))&&qQ(s7(Z,"autonomy/run.sh")))return Z;let Q=rK(Z);if(Q===Z)break;Z=Q}return s7(EO,"..","..","..")}function tK(Z){let X=Z;for(let Q=0;Q<6;Q++){if(qQ(s7(X,"VERSION"))&&qQ(s7(X,"autonomy/run.sh")))return X;let Y=rK(X);if(Y===X)break;X=Y}return s7(Z,"..","..","..")}function j0(){return process.env.LOKI_DIR??s7(process.cwd(),".loki")}function I4(){return s7(d_(),".loki")}var EO,Q8;var G8=p(()=>{EO=rK(p_(import.meta.url));Q8=c_()});import{readFileSync as l_}from"fs";import{resolve as i_,dirname as a_}from"path";import{fileURLToPath as s_}from"url";function _3(){if(f5!==null)return f5;let Z="8.4.0";if(typeof Z==="string"&&Z.length>0)return f5=Z,f5;try{let X=a_(s_(import.meta.url)),Q=tK(X);f5=l_(i_(Q,"VERSION"),"utf-8").trim()}catch{f5="unknown"}return f5}var f5=null;var GQ=p(()=>{G8()});var xO={};c0(xO,{runOrThrow:()=>qf,run:()=>E0,readStreamCapped:()=>HQ,commandVersion:()=>Hf,commandExists:()=>X9,ShellError:()=>eK,MAX_STDOUT_BYTES:()=>kO});async function HQ(Z,X=kO){let Q=Z.getReader(),Y=new TextDecoder,J="",z=0;try{while(z<X){let{done:K,value:$}=await Q.read();if(K)break;if(!$)continue;if(z+=$.byteLength,z>X){let W=$.byteLength-(z-X);J+=Y.decode($.subarray(0,W),{stream:!0});break}J+=Y.decode($,{stream:!0})}J+=Y.decode()}finally{try{await Q.cancel()}catch{}Q.releaseLock()}return J}async function E0(Z,X={}){let Q=Bun.spawn({cmd:[...Z],stdout:"pipe",stderr:"pipe",env:X.env?{...process.env,...X.env}:process.env,cwd:X.cwd}),Y,J;if(X.timeoutMs&&X.timeoutMs>0)Y=setTimeout(()=>{try{Q.kill("SIGTERM")}catch{}J=setTimeout(()=>{try{Q.kill("SIGKILL")}catch{}},2000)},X.timeoutMs);try{let[z,K,$]=await Promise.all([HQ(Q.stdout),new Response(Q.stderr).text(),Q.exited]);return{stdout:z,stderr:K,exitCode:$}}finally{if(Y)clearTimeout(Y);if(J)clearTimeout(J)}}async function qf(Z,X={}){let Q=await E0(Z,X);if(Q.exitCode!==0)throw new eK(`command failed (${Q.exitCode}): ${Z.join(" ")}`,Q.exitCode,Q.stdout,Q.stderr);return Q}async function X9(Z){let X=Gf(Z),Q=await E0(["sh","-c",`command -v ${X}`],{timeoutMs:5000});if(Q.exitCode===0)return Q.stdout.trim()||null;return null}function Gf(Z){if(!/^[A-Za-z0-9._/-]+$/.test(Z))throw Error(`refused to shell-escape suspect token: ${Z}`);return Z}async function Hf(Z,X="--version"){if(!await X9(Z))return null;let Y=await E0([Z,X],{timeoutMs:5000});if(Y.exitCode!==0)return null;return((Y.stdout||Y.stderr).split(/\r?\n/)[0]?.trim()??"")||null}var kO=16777216,eK;var k9=p(()=>{eK=class eK extends Error{message;exitCode;stdout;stderr;constructor(Z,X,Q,Y){super(Z);this.message=Z;this.exitCode=X;this.stdout=Q;this.stderr=Y;this.name="ShellError"}}});function n7(Z){return Uf?"":Z}var Uf,M0,k8,l0,dW0,a0,H8,Q9,g;var S6=p(()=>{Uf=(process.env.NO_COLOR??"").length>0;M0=n7("\x1B[0;31m"),k8=n7("\x1B[0;32m"),l0=n7("\x1B[1;33m"),dW0=n7("\x1B[0;34m"),a0=n7("\x1B[0;36m"),H8=n7("\x1B[1m"),Q9=n7("\x1B[2m"),g=n7("\x1B[0m")});import{existsSync as Ff}from"fs";async function F7(){if(R4!==void 0)return R4;let Z="/opt/homebrew/bin/python3.12";if(Ff(Z))return R4=Z,Z;let X=await X9("python3.12");if(X)return R4=X,X;let Q=await X9("python3");return R4=Q,Q}async function D7(Z,X={}){let Q=await F7();if(!Q)return{stdout:"",stderr:"python3 not found",exitCode:127};return E0([Q,"-c",Z],X)}var R4;var o7=p(()=>{k9()});var rO={};c0(rO,{runStatus:()=>tf});import{existsSync as Y9,readFileSync as h3,readdirSync as dO,statSync as cO}from"fs";import{resolve as h8,basename as pf}from"path";import{homedir as df}from"os";function lO(Z){let X=Math.trunc(Z);if(X>=1e6)return`${(Math.trunc(X/1e6*10)/10).toFixed(1)}M`;if(X>=1000)return`${(Math.trunc(X/1000*10)/10).toFixed(1)}K`;return String(X)}function iO(Z,X,Q){if(X===0)return null;let Y=Math.trunc(Z*100/X),J=Math.trunc(Z*BQ/X);if(J>BQ)J=BQ;let z=BQ-J,K=k8;if(Y>=80)K=M0;else if(Y>=50)K=l0;let $="=".repeat(Math.max(0,J))+" ".repeat(Math.max(0,z)),W=lO(Z),V=lO(X);return`  ${H8}${Q}${g} ${K}[${$}]${g} ${Y}% (${W} / ${V})`}async function lf(){if(await X9("jq"))return!0;return process.stdout.write(`${M0}Error: jq is required but not installed.${g}
+var S_=Object.create;var{getPrototypeOf:y_,defineProperty:oK,getOwnPropertyNames:b_}=Object;var __=Object.prototype.hasOwnProperty;function f_(Z){return this[Z]}var h_,v_,g_=(Z,X,Q)=>{var Y=Z!=null&&typeof Z==="object";if(Y){var J=X?h_??=new WeakMap:v_??=new WeakMap,z=J.get(Z);if(z)return z}Q=Z!=null?S_(y_(Z)):{};let K=X||!Z||!Z.__esModule?oK(Q,"default",{value:Z,enumerable:!0}):Q;for(let $ of b_(Z))if(!__.call(K,$))oK(K,$,{get:f_.bind(Z,$),enumerable:!0});if(Y)J.set(Z,K);return K};var VQ=(Z,X)=>()=>(X||Z((X={exports:{}}).exports,X),X.exports);var m_=(Z)=>Z;function u_(Z,X){this[Z]=m_.bind(null,X)}var c0=(Z,X)=>{for(var Q in X)oK(Z,Q,{get:X[Q],enumerable:!0,configurable:!0,set:u_.bind(X,Q)})};var p=(Z,X)=>()=>(Z&&(X=Z(Z=0)),X);var t0=import.meta.require;var IO={};c0(IO,{lokiDir:()=>j0,homeLokiDir:()=>I4,findRepoRootForVersion:()=>tK,REPO_ROOT:()=>Q8});import{resolve as s7,dirname as rK}from"path";import{fileURLToPath as p_}from"url";import{existsSync as qQ}from"fs";import{homedir as d_}from"os";function c_(){let Z=EO;for(let X=0;X<6;X++){if(qQ(s7(Z,"VERSION"))&&qQ(s7(Z,"autonomy/run.sh")))return Z;let Q=rK(Z);if(Q===Z)break;Z=Q}return s7(EO,"..","..","..")}function tK(Z){let X=Z;for(let Q=0;Q<6;Q++){if(qQ(s7(X,"VERSION"))&&qQ(s7(X,"autonomy/run.sh")))return X;let Y=rK(X);if(Y===X)break;X=Y}return s7(Z,"..","..","..")}function j0(){return process.env.LOKI_DIR??s7(process.cwd(),".loki")}function I4(){return s7(d_(),".loki")}var EO,Q8;var G8=p(()=>{EO=rK(p_(import.meta.url));Q8=c_()});import{readFileSync as l_}from"fs";import{resolve as i_,dirname as a_}from"path";import{fileURLToPath as s_}from"url";function _3(){if(f5!==null)return f5;let Z="8.5.0";if(typeof Z==="string"&&Z.length>0)return f5=Z,f5;try{let X=a_(s_(import.meta.url)),Q=tK(X);f5=l_(i_(Q,"VERSION"),"utf-8").trim()}catch{f5="unknown"}return f5}var f5=null;var GQ=p(()=>{G8()});var xO={};c0(xO,{runOrThrow:()=>qf,run:()=>E0,readStreamCapped:()=>HQ,commandVersion:()=>Hf,commandExists:()=>X9,ShellError:()=>eK,MAX_STDOUT_BYTES:()=>kO});async function HQ(Z,X=kO){let Q=Z.getReader(),Y=new TextDecoder,J="",z=0;try{while(z<X){let{done:K,value:$}=await Q.read();if(K)break;if(!$)continue;if(z+=$.byteLength,z>X){let W=$.byteLength-(z-X);J+=Y.decode($.subarray(0,W),{stream:!0});break}J+=Y.decode($,{stream:!0})}J+=Y.decode()}finally{try{await Q.cancel()}catch{}Q.releaseLock()}return J}async function E0(Z,X={}){let Q=Bun.spawn({cmd:[...Z],stdout:"pipe",stderr:"pipe",env:X.env?{...process.env,...X.env}:process.env,cwd:X.cwd}),Y,J;if(X.timeoutMs&&X.timeoutMs>0)Y=setTimeout(()=>{try{Q.kill("SIGTERM")}catch{}J=setTimeout(()=>{try{Q.kill("SIGKILL")}catch{}},2000)},X.timeoutMs);try{let[z,K,$]=await Promise.all([HQ(Q.stdout),new Response(Q.stderr).text(),Q.exited]);return{stdout:z,stderr:K,exitCode:$}}finally{if(Y)clearTimeout(Y);if(J)clearTimeout(J)}}async function qf(Z,X={}){let Q=await E0(Z,X);if(Q.exitCode!==0)throw new eK(`command failed (${Q.exitCode}): ${Z.join(" ")}`,Q.exitCode,Q.stdout,Q.stderr);return Q}async function X9(Z){let X=Gf(Z),Q=await E0(["sh","-c",`command -v ${X}`],{timeoutMs:5000});if(Q.exitCode===0)return Q.stdout.trim()||null;return null}function Gf(Z){if(!/^[A-Za-z0-9._/-]+$/.test(Z))throw Error(`refused to shell-escape suspect token: ${Z}`);return Z}async function Hf(Z,X="--version"){if(!await X9(Z))return null;let Y=await E0([Z,X],{timeoutMs:5000});if(Y.exitCode!==0)return null;return((Y.stdout||Y.stderr).split(/\r?\n/)[0]?.trim()??"")||null}var kO=16777216,eK;var k9=p(()=>{eK=class eK extends Error{message;exitCode;stdout;stderr;constructor(Z,X,Q,Y){super(Z);this.message=Z;this.exitCode=X;this.stdout=Q;this.stderr=Y;this.name="ShellError"}}});function n7(Z){return Uf?"":Z}var Uf,M0,k8,l0,dW0,a0,H8,Q9,g;var S6=p(()=>{Uf=(process.env.NO_COLOR??"").length>0;M0=n7("\x1B[0;31m"),k8=n7("\x1B[0;32m"),l0=n7("\x1B[1;33m"),dW0=n7("\x1B[0;34m"),a0=n7("\x1B[0;36m"),H8=n7("\x1B[1m"),Q9=n7("\x1B[2m"),g=n7("\x1B[0m")});import{existsSync as Ff}from"fs";async function F7(){if(R4!==void 0)return R4;let Z="/opt/homebrew/bin/python3.12";if(Ff(Z))return R4=Z,Z;let X=await X9("python3.12");if(X)return R4=X,X;let Q=await X9("python3");return R4=Q,Q}async function D7(Z,X={}){let Q=await F7();if(!Q)return{stdout:"",stderr:"python3 not found",exitCode:127};return E0([Q,"-c",Z],X)}var R4;var o7=p(()=>{k9()});var rO={};c0(rO,{runStatus:()=>tf});import{existsSync as Y9,readFileSync as h3,readdirSync as dO,statSync as cO}from"fs";import{resolve as h8,basename as pf}from"path";import{homedir as df}from"os";function lO(Z){let X=Math.trunc(Z);if(X>=1e6)return`${(Math.trunc(X/1e6*10)/10).toFixed(1)}M`;if(X>=1000)return`${(Math.trunc(X/1000*10)/10).toFixed(1)}K`;return String(X)}function iO(Z,X,Q){if(X===0)return null;let Y=Math.trunc(Z*100/X),J=Math.trunc(Z*BQ/X);if(J>BQ)J=BQ;let z=BQ-J,K=k8;if(Y>=80)K=M0;else if(Y>=50)K=l0;let $="=".repeat(Math.max(0,J))+" ".repeat(Math.max(0,z)),W=lO(Z),V=lO(X);return`  ${H8}${Q}${g} ${K}[${$}]${g} ${Y}% (${W} / ${V})`}async function lf(){if(await X9("jq"))return!0;return process.stdout.write(`${M0}Error: jq is required but not installed.${g}
 `),process.stdout.write(`Install with:
 `),process.stdout.write(`  brew install jq    (macOS)
 `),process.stdout.write(`  apt install jq     (Debian/Ubuntu)
@@ -1206,4 +1206,4 @@ Set LOKI_LEGACY_BASH=1 to force the bash CLI for every command.
 `),2}case"start":{let{runStart:Y}=await Promise.resolve().then(() => (R_(),P_));return Y(Q)}default:return process.stderr.write(`Unknown command: ${X}
 `),process.stderr.write(k_),2}}uO();process.on("SIGINT",()=>process.exit(130));process.on("SIGTERM",()=>process.exit(143));var DW0=await FW0(Bun.argv.slice(2));process.exit(DW0);
 
-//# debugId=8285CFE544C8109D64756E2164756E21
+//# debugId=FFE6617DCF7BFED164756E2164756E21

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "loki-mode",
   "mcpName": "io.github.asklokesh/loki-mode",
-  "version": "8.4.0",
+  "version": "8.5.0",
   "description": "Loki Mode by Autonomi. Autonomous spec-to-product system: takes a PRD, GitHub issue, OpenAPI/JSON/YAML, or one-line brief to a deployed app via the RARV-C closure loop with 8 quality gates. Provider-agnostic (Claude Code, OpenAI Codex, Cline, Aider).",
   "keywords": [
     "agent",

package/plugins/loki-mode/.claude-plugin/plugin.json

@@ -2,7 +2,7 @@
   "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
   "name": "loki-mode",
   "displayName": "Loki Mode",
-  "version": "8.4.0",
+  "version": "8.5.0",
   "description": "Autonomous spec-to-product build system with a built-in trust layer (RARV-C closure loop, 8 quality gates, completion council). Ships Loki's spec-hardening, drift-detection, and deterministic PR verification commands plus the Loki MCP server.",
   "author": {
     "name": "Autonomi",