loki-mode 8.42.0 → 8.44.0

package/SKILL.md

@@ -3,7 +3,7 @@ name: loki-mode
 description: Autonomous spec-driven build system with a built-in trust layer. It does not call work done until it is verified (RARV-C closure loop, 8 quality gates, completion council, verified-completion evidence gate). Triggers on "Loki Mode". Takes a spec (PRD, GitHub issue, OpenAPI doc, etc.) to deployed product with minimal human intervention. Provider-agnostic. Requires --dangerously-skip-permissions flag.
 ---
 
-# Loki Mode v8.42.0
+# Loki Mode v8.44.0
 
 **You are an autonomous agent. You make decisions. You do not ask questions. You do not stop.**
 
@@ -469,4 +469,4 @@ See `CHANGELOG.md` entries [7.5.7], [7.5.8], [7.5.13] for the per-fix list and r
 
 ---
 
-**v8.42.0 | [Autonomi](https://www.autonomi.dev/) flagship product | ~410 lines core**
+**v8.44.0 | [Autonomi](https://www.autonomi.dev/) flagship product | ~410 lines core**

package/VERSION

@@ -1 +1 @@
-8.42.0
+8.44.0

package/autonomy/loki

@@ -11626,6 +11626,38 @@ else:
     fi
     echo ""
 
+    # INSTALL INTEGRITY: the quality gates shell out to detector scripts. If the
+    # package shipped without them, four gates fail-CLOSED on every iteration --
+    # correct behaviour (a missing detector must never read as "nothing found"),
+    # but it means the build can never pass and first-pass completion is
+    # impossible no matter how good the output is.
+    #
+    # That shipped: package.json's files[] had no tests/ entry until v8.38.0, so
+    # every npm install had zero detectors. `doctor` reported a healthy system
+    # the whole time, because it only ever checked for external COMMANDS -- not
+    # for the files this package is supposed to contain. A broken install and a
+    # healthy one were indistinguishable from the one command meant to tell them
+    # apart.
+    echo -e "${BOLD}Install integrity:${NC}"
+    local _det_missing="" _det
+    for _det in detect-test-mutations detect-mock-problems \
+                detect-semantic-test-problems detect-invariant-violations; do
+        if [ -f "${_LOKI_SCRIPT_DIR}/../tests/${_det}.sh" ]; then
+            pass_count=$((pass_count + 1))
+        else
+            _det_missing="${_det_missing} ${_det}.sh"
+        fi
+    done
+    if [ -z "$_det_missing" ]; then
+        echo -e "  ${GREEN}OK${NC}    Quality-gate detectors present (4/4)"
+    else
+        echo -e "  ${RED}FAIL${NC}  Quality-gate detectors MISSING:${_det_missing}"
+        echo -e "  ${DIM}      These gates fail-closed, so every iteration will be blocked.${NC}"
+        _doctor_block "Incomplete install: quality-gate detectors are missing. Reinstall: bun install -g loki-mode"
+        fail_count=$((fail_count + 1))
+    fi
+    echo ""
+
     # Summary
     echo -e "${BOLD}Summary:${NC} ${GREEN}$pass_count passed${NC}, ${RED}$fail_count failed${NC}, ${YELLOW}$warn_count warnings${NC}"
     echo ""

package/autonomy/run.sh

@@ -23036,8 +23036,30 @@ if __name__ == "__main__":
                     # v7.5.3 Phase 1 hook: persist structured findings +
                     # auto-write learnings (one shell-out per iteration).
                     # Best-effort; never fails the main loop.
-                    if [ "${LOKI_INJECT_FINDINGS:-1}" != "0" ] && command -v bun >/dev/null 2>&1; then
-                        bun "${SCRIPT_DIR}/../loki-ts/dist/loki.js" internal phase1-hooks reflect "$ITERATION_COUNT" 2>/dev/null || true
+                    if [ "${LOKI_INJECT_FINDINGS:-1}" != "0" ]; then
+                        if command -v bun >/dev/null 2>&1; then
+                            bun "${SCRIPT_DIR}/../loki-ts/dist/loki.js" internal phase1-hooks reflect "$ITERATION_COUNT" 2>/dev/null || true
+                        else
+                            # DEGRADED, and said so. Findings injection is what
+                            # tells the next iteration WHAT to fix; without it the
+                            # agent knows only that it failed. Research puts "the
+                            # agent did not attempt to recover from an error" at
+                            # 56% of all agent failures, and a feedback loop that
+                            # silently stops feeding back manufactures exactly that
+                            # shape -- the next iteration then looks like the model
+                            # failing, when it was never told what went wrong.
+                            #
+                            # Defaults ON but was gated on `command -v bun`, so on a
+                            # machine without bun it degraded with no signal at all.
+                            # A missing capability must be visible; a silent one is
+                            # worse than an absent feature because it misattributes
+                            # the failure.
+                            log_warn "Findings injection unavailable (bun not found): the next iteration will be told it failed but NOT what to fix. Install bun, or set LOKI_INJECT_FINDINGS=0 to silence this."
+                            emit_event_json "capability_degraded" \
+                                "capability=inject_findings" \
+                                "reason=bun_not_found" \
+                                "impact=next_iteration_lacks_structured_findings" 2>/dev/null || true
+                        fi
                     fi
                 fi
                 emit_stage_complete "code_review" "$_stg_ok" "$_stg_t0"

package/dashboard/__init__.py

@@ -7,7 +7,7 @@ Modules:
     control: Session control API (start/stop/pause/resume)
 """
 
-__version__ = "8.42.0"
+__version__ = "8.44.0"
 
 # Expose the control app for easy import
 try:

package/docs/FIRST-PASS-COMPLETION-PLAN.md

@@ -0,0 +1,161 @@
+# Fastest First-Pass Full Completion
+
+Founder ask: **fastest first-pass full completion.** No waiting through a second
+iteration. Highest quality output. Grounded in researched competitor data, not
+test auditing.
+
+This replaces the speed framing of `SPEED-AND-FIRST-PASS-PLAN.md`. That plan
+optimised the *duration* of iterations. This one attacks the *number* of them,
+which is what the founder actually asked for and where the evidence points.
+
+---
+
+## 1. The research says the bottleneck is not the model
+
+| finding | number | source |
+|---|---|---|
+| Agent failures traced to **planning, before the first line of code** | **82%** | loadsys 2026 failure-rate analysis |
+| First-iteration aggregate pass rate | **47.8%** | SlopCodeBench (arXiv 2603.24755) |
+| Failures where the agent **did not attempt to recover** from an error | **56%** | same |
+| Visible resolutions **requiring explicit user correction** | **91.49%** | 20,574-session misalignment study (arXiv 2605.29442) |
+| Typical iteration cycles per task | **3-5** | Anthropic SWE-bench internal |
+| Cursor Agent Loop default iteration cap | **8** | Cursor 2026 breakdown |
+
+Two conclusions, and they point the same way:
+
+1. **Second iterations are caused upstream of the model.** 82% of failures are
+   already determined before code is written. Making the model faster or the
+   council shorter cannot fix a task that was mis-specified.
+2. **Nobody has solved this.** Cursor ships an 8-iteration cap. Devin resolves
+   51.5% of issues. The category *assumes* iteration. First-pass completion is
+   an unclaimed position, not a race we are losing.
+
+**Strategic consequence.** We cannot out-model Cursor (custom MoE, ~250 tok/s)
+or Cognition (SWE-1.6, ~950 tok/s). We do not have to. The measured
+bottleneck -- specification quality and error recovery -- is harness work, and
+harness work is the thing we can actually do.
+
+## 2. Our own data says exactly the same thing
+
+Every recorded run on this machine:
+
+| project | iterations | gates that failed |
+|---|---:|---|
+| FireLater | 3 | static_analysis 2, **mutation_integrity 3**, code_review 2 |
+| anonima | 4 | code_review 1 |
+| autonomi-engine-runs | 1 | none |
+| loki-mode | 1 | none |
+
+**Perfect correlation: every multi-iteration run had a failing gate; every
+single-iteration run had none.** Iterations are not the model failing to
+finish. They are gates rejecting work and sending it back.
+
+That reframes the goal precisely:
+
+> **First-pass completion = the first iteration passes every gate.**
+> Not "the model tries harder". Not "fewer gates".
+
+And it explains the founder's 21-minute FireLater run better than any latency
+measurement: `mutation_integrity` failed **3 of 3 iterations in 0-1 seconds
+each** -- the detector was never packaged (fixed v8.38.0), so the gate could
+never pass. **First-pass completion was arithmetically impossible.** No speed
+work would have touched it.
+
+## 3. The plan
+
+Ranked by measured contribution to iteration count.
+
+### F0 -- A gate that cannot pass must abort, not iterate (highest value)
+
+FireLater burned 3 iterations against a gate whose detector did not exist. The
+gate correctly fail-closed each time; nothing noticed it was failing for the
+same unfixable reason.
+
+- Detect a gate failing with an **identical cause** across iterations and stop
+  with a named terminal reason instead of re-running the model.
+- This is the 56% "did not attempt to recover" bucket, inverted: when recovery
+  is *impossible*, the honest move is to say so in iteration 1.
+- **Already half-built:** `#87` does exactly this for spec contradictions
+  (fail FAST + HONEST + NAMED rather than grinding to max-iterations). Extend
+  that proven pattern to unfixable gate failures.
+- Guard rails: only on a byte-identical repeated cause, never a first failure,
+  and it must map to a terminal-failure exit -- never a fake green.
+
+### F1 -- Front-load the 82%: make the spec gate the first-pass gate
+
+`LOKI_SPEC_GRILL` already interrogates the spec before the loop and defaults
+ON. That is the correct lever for the 82% planning bucket, and it is already
+paid for.
+
+- Measure what it actually catches per run (it is unmeasured today).
+- Extend interrogation from *contradictions* to the ambiguity classes the
+  20,574-session study names: unstated acceptance criteria, unstated scope
+  bounds, unstated interaction contracts.
+- **Research constraint, load-bearing:** auto-generated context files measured
+  **-3% success, +20% cost**; human-written ones **+4%**. So this must produce
+  *questions and resolutions*, never a generated context blob.
+
+### F2 -- Findings injection must not be silently optional
+
+`LOKI_INJECT_FINDINGS` defaults on, but the injection is gated on
+`command -v bun`. **Without bun, the agent is told it failed and not what to
+fix** -- the exact "no error recovery" shape that is 56% of failures.
+
+- Make the fallback explicit: if findings cannot be injected, say so loudly.
+- A silent degradation of the feedback loop is worse than a missing feature,
+  because the next iteration looks like a model failure.
+
+### F3 -- Verify before the gate, not after
+
+47.8% first-iteration pass rate is the industry number. The cheap deterministic
+gates (test_suite, static_analysis, lsp_diagnostics) cost ~6s combined and run
+*after* the model has already declared done.
+
+- Have the agent run them **during** iteration 1 and fix what they report,
+  before the iteration closes.
+- Cursor's Agent Loop is exactly this: "runs the test suite, reads stderr,
+  edits the offending files, re-runs" -- their headline architectural change.
+  We already have the gates; we just run them too late to help pass 1.
+
+### F4 -- Iteration budget as a measured decision
+
+Research: 1-2 iteration caps fail even when the approach was sound; 5-10 is the
+recommended range. We ship `LOKI_MAX_ITERATIONS=1000`.
+
+- The goal is not a small cap. It is **finishing in one** and stopping honestly
+  when one is not enough.
+- Pairs with F0: a cap is a blunt instrument; a named terminal reason is a
+  diagnosis.
+
+## 4. What we do NOT do
+
+- **No model training.** Cursor and Cognition bought speed with custom models.
+  Not reachable, and claiming otherwise would be dishonest.
+- **No generated context files.** Measured -3% success, +20% cost.
+- **No weakening of gates to raise first-pass rate.** A gate that stops
+  blocking does not improve completion; it fakes it. Every item above either
+  fixes the *cause* of a failure or reports it honestly.
+
+## 5. Acceptance
+
+The measurement is already built (`scripts/measure-run.sh`, v8.37.0):
+
+> a scoped GitHub issue completes in **one iteration**, with every gate
+> genuinely passing, and when it cannot, the run stops in iteration 1 with a
+> named reason instead of grinding.
+
+Baseline to beat, from our own table: FireLater 3 iterations, anonima 4.
+
+## 6. Why this is a real edge
+
+The category has conceded iteration -- Cursor caps it at 8, Devin checkpoints
+through it. Nobody markets first-pass completion because nobody has it.
+
+Our moat is the Evidence Receipt: we can *prove* an iteration passed every
+gate. Combining that with genuine first-pass completion is a claim no
+competitor can make, and it is reachable with harness work rather than a
+frontier lab.
+
+Sources: SlopCodeBench (arXiv 2603.24755), developer-agent misalignment study
+(arXiv 2605.29442), loadsys 2026 agent failure-rate analysis, Cursor 2026 agent
+loop breakdown, Devin SWE-bench reporting.

package/loki-ts/dist/loki.js

@@ -1,5 +1,5 @@
 // @bun
-var h_=Object.create;var{getPrototypeOf:v_,defineProperty:rK,getOwnPropertyNames:g_}=Object;var m_=Object.prototype.hasOwnProperty;function u_(Z){return this[Z]}var p_,d_,c_=(Z,X,Q)=>{var Y=Z!=null&&typeof Z==="object";if(Y){var J=X?p_??=new WeakMap:d_??=new WeakMap,z=J.get(Z);if(z)return z}Q=Z!=null?h_(v_(Z)):{};let K=X||!Z||!Z.__esModule?rK(Q,"default",{value:Z,enumerable:!0}):Q;for(let $ of g_(Z))if(!m_.call(K,$))rK(K,$,{get:u_.bind(Z,$),enumerable:!0});if(Y)J.set(Z,K);return K};var qQ=(Z,X)=>()=>(X||Z((X={exports:{}}).exports,X),X.exports);var l_=(Z)=>Z;function i_(Z,X){this[Z]=l_.bind(null,X)}var l0=(Z,X)=>{for(var Q in X)rK(Z,Q,{get:X[Q],enumerable:!0,configurable:!0,set:i_.bind(X,Q)})};var p=(Z,X)=>()=>(Z&&(X=Z(Z=0)),X);var e0=import.meta.require;var kO={};l0(kO,{lokiDir:()=>j0,homeLokiDir:()=>P4,findRepoRootForVersion:()=>eK,REPO_ROOT:()=>r0});import{resolve as n7,dirname as tK}from"path";import{fileURLToPath as a_}from"url";import{existsSync as GQ}from"fs";import{homedir as s_}from"os";function n_(){let Z=RO;for(let X=0;X<6;X++){if(GQ(n7(Z,"VERSION"))&&GQ(n7(Z,"autonomy/run.sh")))return Z;let Q=tK(Z);if(Q===Z)break;Z=Q}return n7(RO,"..","..","..")}function eK(Z){let X=Z;for(let Q=0;Q<6;Q++){if(GQ(n7(X,"VERSION"))&&GQ(n7(X,"autonomy/run.sh")))return X;let Y=tK(X);if(Y===X)break;X=Y}return n7(Z,"..","..","..")}function j0(){return process.env.LOKI_DIR??n7(process.cwd(),".loki")}function P4(){return n7(s_(),".loki")}var RO,r0;var G8=p(()=>{RO=tK(a_(import.meta.url));r0=n_()});import{readFileSync as o_}from"fs";import{resolve as r_,dirname as t_}from"path";import{fileURLToPath as e_}from"url";function f3(){if(h5!==null)return h5;let Z="8.42.0";if(typeof Z==="string"&&Z.length>0)return h5=Z,h5;try{let X=t_(e_(import.meta.url)),Q=eK(X);h5=o_(r_(Q,"VERSION"),"utf-8").trim()}catch{h5="unknown"}return h5}var h5=null;var HQ=p(()=>{G8()});var bO={};l0(bO,{runOrThrow:()=>Of,run:()=>E0,readStreamCapped:()=>UQ,commandVersion:()=>Af,commandExists:()=>X9,ShellError:()=>Z$,MAX_STDOUT_BYTES:()=>yO});async function UQ(Z,X=yO){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,$]=await Promise.all([UQ(Q.stdout),new Response(Q.stderr).text(),Q.exited]);return{stdout:z,stderr:K,exitCode:$}}finally{if(Y)clearTimeout(Y);if(J)clearTimeout(J)}}async function Of(Z,X={}){let Q=await E0(Z,X);if(Q.exitCode!==0)throw new Z$(`command failed (${Q.exitCode}): ${Z.join(" ")}`,Q.exitCode,Q.stdout,Q.stderr);return Q}async function X9(Z){let X=Lf(Z),Q=await E0(["sh","-c",`command -v ${X}`],{timeoutMs:5000});if(Q.exitCode===0)return Q.stdout.trim()||null;return null}function Lf(Z){if(!/^[A-Za-z0-9._/-]+$/.test(Z))throw Error(`refused to shell-escape suspect token: ${Z}`);return Z}async function Af(Z,X="--version"){if(!await X9(Z))return null;let Y=await E0([Z,X],{timeoutMs:5000});if(Y.exitCode!==0)return null;return((Y.stdout||Y.stderr).split(/\r?\n/)[0]?.trim()??"")||null}var yO=16777216,Z$;var x9=p(()=>{Z$=class Z$ 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 o7(Z){return jf?"":Z}var jf,L0,k8,p0,ZV0,i0,H8,Q9,v;var S6=p(()=>{jf=(process.env.NO_COLOR??"").length>0;L0=o7("\x1B[0;31m"),k8=o7("\x1B[0;32m"),p0=o7("\x1B[1;33m"),ZV0=o7("\x1B[0;34m"),i0=o7("\x1B[0;36m"),H8=o7("\x1B[1m"),Q9=o7("\x1B[2m"),v=o7("\x1B[0m")});import{existsSync as xf}from"fs";async function E7(){if(k4!==void 0)return k4;let Z="/opt/homebrew/bin/python3.12";if(xf(Z))return k4=Z,Z;let X=await X9("python3.12");if(X)return k4=X,X;let Q=await X9("python3");return k4=Q,Q}async function Y7(Z,X={}){let Q=await E7();if(!Q)return{stdout:"",stderr:"python3 not found",exitCode:127};return E0([Q,"-c",Z],X)}var k4;var r7=p(()=>{x9()});var ZL={};l0(ZL,{runStatus:()=>Yh});import{existsSync as Y9,readFileSync as v3,readdirSync as iO,statSync as aO}from"fs";import{resolve as h8,basename as sf}from"path";import{homedir as nf}from"os";function sO(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 nO(Z,X,Q){if(X===0)return null;let Y=Math.trunc(Z*100/X),J=Math.trunc(Z*NQ/X);if(J>NQ)J=NQ;let z=NQ-J,K=k8;if(Y>=80)K=L0;else if(Y>=50)K=p0;let $="=".repeat(Math.max(0,J))+" ".repeat(Math.max(0,z)),W=sO(Z),V=sO(X);return`  ${H8}${Q}${v} ${K}[${$}]${v} ${Y}% (${W} / ${V})`}async function rf(){if(await X9("jq"))return!0;return process.stdout.write(`${L0}Error: jq is required but not installed.${v}
+var h_=Object.create;var{getPrototypeOf:v_,defineProperty:rK,getOwnPropertyNames:g_}=Object;var m_=Object.prototype.hasOwnProperty;function u_(Z){return this[Z]}var p_,d_,c_=(Z,X,Q)=>{var Y=Z!=null&&typeof Z==="object";if(Y){var J=X?p_??=new WeakMap:d_??=new WeakMap,z=J.get(Z);if(z)return z}Q=Z!=null?h_(v_(Z)):{};let K=X||!Z||!Z.__esModule?rK(Q,"default",{value:Z,enumerable:!0}):Q;for(let $ of g_(Z))if(!m_.call(K,$))rK(K,$,{get:u_.bind(Z,$),enumerable:!0});if(Y)J.set(Z,K);return K};var qQ=(Z,X)=>()=>(X||Z((X={exports:{}}).exports,X),X.exports);var l_=(Z)=>Z;function i_(Z,X){this[Z]=l_.bind(null,X)}var l0=(Z,X)=>{for(var Q in X)rK(Z,Q,{get:X[Q],enumerable:!0,configurable:!0,set:i_.bind(X,Q)})};var p=(Z,X)=>()=>(Z&&(X=Z(Z=0)),X);var e0=import.meta.require;var kO={};l0(kO,{lokiDir:()=>j0,homeLokiDir:()=>P4,findRepoRootForVersion:()=>eK,REPO_ROOT:()=>r0});import{resolve as n7,dirname as tK}from"path";import{fileURLToPath as a_}from"url";import{existsSync as GQ}from"fs";import{homedir as s_}from"os";function n_(){let Z=RO;for(let X=0;X<6;X++){if(GQ(n7(Z,"VERSION"))&&GQ(n7(Z,"autonomy/run.sh")))return Z;let Q=tK(Z);if(Q===Z)break;Z=Q}return n7(RO,"..","..","..")}function eK(Z){let X=Z;for(let Q=0;Q<6;Q++){if(GQ(n7(X,"VERSION"))&&GQ(n7(X,"autonomy/run.sh")))return X;let Y=tK(X);if(Y===X)break;X=Y}return n7(Z,"..","..","..")}function j0(){return process.env.LOKI_DIR??n7(process.cwd(),".loki")}function P4(){return n7(s_(),".loki")}var RO,r0;var G8=p(()=>{RO=tK(a_(import.meta.url));r0=n_()});import{readFileSync as o_}from"fs";import{resolve as r_,dirname as t_}from"path";import{fileURLToPath as e_}from"url";function f3(){if(h5!==null)return h5;let Z="8.44.0";if(typeof Z==="string"&&Z.length>0)return h5=Z,h5;try{let X=t_(e_(import.meta.url)),Q=eK(X);h5=o_(r_(Q,"VERSION"),"utf-8").trim()}catch{h5="unknown"}return h5}var h5=null;var HQ=p(()=>{G8()});var bO={};l0(bO,{runOrThrow:()=>Of,run:()=>E0,readStreamCapped:()=>UQ,commandVersion:()=>Af,commandExists:()=>X9,ShellError:()=>Z$,MAX_STDOUT_BYTES:()=>yO});async function UQ(Z,X=yO){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,$]=await Promise.all([UQ(Q.stdout),new Response(Q.stderr).text(),Q.exited]);return{stdout:z,stderr:K,exitCode:$}}finally{if(Y)clearTimeout(Y);if(J)clearTimeout(J)}}async function Of(Z,X={}){let Q=await E0(Z,X);if(Q.exitCode!==0)throw new Z$(`command failed (${Q.exitCode}): ${Z.join(" ")}`,Q.exitCode,Q.stdout,Q.stderr);return Q}async function X9(Z){let X=Lf(Z),Q=await E0(["sh","-c",`command -v ${X}`],{timeoutMs:5000});if(Q.exitCode===0)return Q.stdout.trim()||null;return null}function Lf(Z){if(!/^[A-Za-z0-9._/-]+$/.test(Z))throw Error(`refused to shell-escape suspect token: ${Z}`);return Z}async function Af(Z,X="--version"){if(!await X9(Z))return null;let Y=await E0([Z,X],{timeoutMs:5000});if(Y.exitCode!==0)return null;return((Y.stdout||Y.stderr).split(/\r?\n/)[0]?.trim()??"")||null}var yO=16777216,Z$;var x9=p(()=>{Z$=class Z$ 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 o7(Z){return jf?"":Z}var jf,L0,k8,p0,ZV0,i0,H8,Q9,v;var S6=p(()=>{jf=(process.env.NO_COLOR??"").length>0;L0=o7("\x1B[0;31m"),k8=o7("\x1B[0;32m"),p0=o7("\x1B[1;33m"),ZV0=o7("\x1B[0;34m"),i0=o7("\x1B[0;36m"),H8=o7("\x1B[1m"),Q9=o7("\x1B[2m"),v=o7("\x1B[0m")});import{existsSync as xf}from"fs";async function E7(){if(k4!==void 0)return k4;let Z="/opt/homebrew/bin/python3.12";if(xf(Z))return k4=Z,Z;let X=await X9("python3.12");if(X)return k4=X,X;let Q=await X9("python3");return k4=Q,Q}async function Y7(Z,X={}){let Q=await E7();if(!Q)return{stdout:"",stderr:"python3 not found",exitCode:127};return E0([Q,"-c",Z],X)}var k4;var r7=p(()=>{x9()});var ZL={};l0(ZL,{runStatus:()=>Yh});import{existsSync as Y9,readFileSync as v3,readdirSync as iO,statSync as aO}from"fs";import{resolve as h8,basename as sf}from"path";import{homedir as nf}from"os";function sO(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 nO(Z,X,Q){if(X===0)return null;let Y=Math.trunc(Z*100/X),J=Math.trunc(Z*NQ/X);if(J>NQ)J=NQ;let z=NQ-J,K=k8;if(Y>=80)K=L0;else if(Y>=50)K=p0;let $="=".repeat(Math.max(0,J))+" ".repeat(Math.max(0,z)),W=sO(Z),V=sO(X);return`  ${H8}${Q}${v} ${K}[${$}]${v} ${Y}% (${W} / ${V})`}async function rf(){if(await X9("jq"))return!0;return process.stdout.write(`${L0}Error: jq is required but not installed.${v}
 `),process.stdout.write(`Install with:
 `),process.stdout.write(`  brew install jq    (macOS)
 `),process.stdout.write(`  apt install jq     (Debian/Ubuntu)
@@ -1222,4 +1222,4 @@ Set LOKI_LEGACY_BASH=1 to force the bash CLI for every command.
 `),2}case"start":{let{runStart:Y}=await Promise.resolve().then(() => (b_(),y_));return Y(Q)}default:return process.stderr.write(`Unknown command: ${X}
 `),process.stderr.write(__),2}}cO();process.on("SIGINT",()=>process.exit(130));process.on("SIGTERM",()=>process.exit(143));var fW0=await _W0(Bun.argv.slice(2));process.exit(fW0);
 
-//# debugId=FF2C6FF7ACDA238764756E2164756E21
+//# debugId=E6F40E8D30865CF164756E2164756E21

package/mcp/__init__.py

@@ -75,4 +75,4 @@ try:
 except ImportError:
     __all__ = ['mcp']
 
-__version__ = '8.42.0'
+__version__ = '8.44.0'

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "loki-mode",
   "mcpName": "io.github.asklokesh/loki-mode",
-  "version": "8.42.0",
+  "version": "8.44.0",
   "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": "8.42.0",
+  "version": "8.44.0",
   "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",