npm - @codewithjuber/forgekit - Versions diffs - 0.8.0 → 0.8.1 - Mend

@codewithjuber/forgekit 0.8.0 → 0.8.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (17) hide show

package/.claude-plugin/plugin.json +1 -1
package/.codex-plugin/plugin.json +2 -2
package/CHANGELOG.md +6 -1
package/package.json +1 -2
package/public/index.html +3 -3
package/src/cli.js +30 -7
package/src/cortex_mcp.js +74 -1
package/src/dash.html +31 -24
package/src/dash.js +36 -0
package/src/emit/copilot.js +1 -1
package/src/emit/gemini.js +1 -1
package/src/emit/mcp.js +2 -2
package/src/init.js +15 -4
package/src/sync.js +1 -1
package/templates/project-layer/CLAUDE.md +11 -14
package/skills/cognitive-substrate/SKILL.md +0 -56
package/skills/cognitive-substrate/references/capability-map.md +0 -17

package/.claude-plugin/plugin.json CHANGED Viewed

@@ -2,7 +2,7 @@
   "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
   "name": "forgekit",
   "displayName": "Forge",
-  "version": "0.8.0",
+  "version": "0.8.1",
   "description": "One config, every AI coding tool — cognitive substrate, tools, crew, guards, atlas, lean, recall from one source.",
   "author": { "name": "CodeWithJuber" },
   "license": "MIT",

package/.codex-plugin/plugin.json CHANGED Viewed

@@ -1,13 +1,13 @@
 {
   "name": "forgekit",
-  "version": "0.8.0",
+  "version": "0.8.1",
   "description": "One config, every AI coding tool — cognitive substrate, MCP tools, guards, atlas, recall, and routing from one source.",
   "author": { "name": "CodeWithJuber", "url": "https://github.com/CodeWithJuber" },
   "homepage": "https://github.com/CodeWithJuber/forgekit#readme",
   "repository": "https://github.com/CodeWithJuber/forgekit",
   "license": "MIT",
   "keywords": ["codex", "mcp", "ai-coding", "agents-md", "cognitive-substrate"],
-  "skills": "skills",
+  "skills": "global/tools",
   "mcpServers": ".mcp.json",
   "interface": {
     "displayName": "Forge",

package/CHANGELOG.md CHANGED Viewed

@@ -6,6 +6,10 @@ to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 ## [Unreleased]
+## [0.8.1] - 2026-07-08
 ## [0.8.0] - 2026-07-08
@@ -454,7 +458,8 @@ to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
   check; coverage + type-checking (`tsc --checkJs`); 2026 production-standard rules;
   OWASP-LLM / NIST SSDF / SLSA control mapping.
-[Unreleased]: https://github.com/CodeWithJuber/forgekit/compare/v0.8.0...HEAD
+[Unreleased]: https://github.com/CodeWithJuber/forgekit/compare/v0.8.1...HEAD
+[0.8.1]: https://github.com/CodeWithJuber/forgekit/compare/v0.8.0...v0.8.1
 [0.8.0]: https://github.com/CodeWithJuber/forgekit/compare/v0.7.0...v0.8.0
 [0.7.0]: https://github.com/CodeWithJuber/forgekit/compare/v0.6.0...v0.7.0
 [0.6.0]: https://github.com/CodeWithJuber/forgekit/compare/v0.5.0...v0.6.0

package/package.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "name": "@codewithjuber/forgekit",
-  "version": "0.8.0",
+  "version": "0.8.1",
   "description": "One brain for every AI coding agent \u2014 the cognitive substrate every frozen model is missing (proof-carrying memory, impact foresight, enforced guardrails), authored once and delivered as native config to Claude Code, Codex, Cursor, Gemini, Aider, and more.",
   "type": "module",
   "bin": {
@@ -39,7 +39,6 @@
     "hooks",
     ".claude-plugin",
     ".codex-plugin",
-    "skills",
     ".mcp.json",
     "bin",
     "brand.json",

package/public/index.html CHANGED Viewed

@@ -1,4 +1,4 @@
-<!doctype html><html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1"><title>forgekit status — live repository data</title><meta name="description" content="One config for every AI coding agent — a cross-tool config layer plus a cognitive substrate (memory, blast-radius, guardrails) for Claude Code, Codex, Cursor, Gemini, Aider, and more."><meta name="theme-color" content="#171310"><style>
+<!doctype html><html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1"><title>forgekit status — live repository data</title><meta name="description" content="One brain for every AI coding agent — the cognitive substrate every frozen model is missing (proof-carrying memory, impact foresight, enforced guardrails), authored once and delivered as native config to Claude Code, Codex, Cursor, Gemini, Aider, and more."><meta name="theme-color" content="#171310"><style>
 :root{--ink:#171310;--surface:#201a15;--surface-2:#272019;--line:#372c22;--faint:#7d7263;--muted:#a99e90;--ember:#f26430;--fg:#f2ede7;--r-s:6px;--r-m:12px;--r-pill:999px;--shadow:0 24px 64px rgba(23,19,16,.56);--mono:ui-monospace,"SF Mono",SFMono-Regular,Menlo,Consolas,monospace;--sans:system-ui,-apple-system,"Segoe UI",Roboto,sans-serif}
 *{box-sizing:border-box}
 body{margin:0;background:radial-gradient(960px 400px at 85% -10%,rgba(242,100,48,.16),transparent 70%),var(--ink);color:var(--fg);font:16px/1.65 var(--sans);-webkit-font-smoothing:antialiased}
@@ -39,7 +39,7 @@ h2{font-size:24px;letter-spacing:-.01em;margin:0 0 16px}
 code{font:500 13px var(--mono);background:var(--surface-2);border:1px solid var(--line);border-radius:var(--r-s);padding:0 8px}
 footer{padding:32px 0;color:var(--faint);font-size:14px}
 @media(max-width:800px){.grid{grid-template-columns:1fr}}
-</style></head><body><div class="wrap"><header class="nav"><a class="brand" href="../">forge<em>kit</em></a><nav class="links" aria-label="Primary"><a href="../">Landing</a><a href="#quickstart">Quickstart</a><a href="#changes">Latest</a><a href="#sources">Data Sources</a><a href="https://github.com/CodeWithJuber/forgekit">GitHub ↗</a></nav></header><main id="top"><section class="hero" style="border-bottom:0;padding-bottom:0"><p class="eyebrow">@codewithjuber/forgekit · v0.6.0 · Node &gt;=20</p><h1>Live status, straight from the repository.</h1><p class="lead">One config for every AI coding agent — a cross-tool config layer plus a cognitive substrate (memory, blast-radius, guardrails) for Claude Code, Codex, Cursor, Gemini, Aider, and more.</p><p><a class="btn primary" href="#quickstart">Install in 60 seconds</a> <a class="btn" href="https://github.com/CodeWithJuber/forgekit#readme">Read the docs</a></p><div class="meta"><span class="chip">MIT license</span><span class="chip">0 runtime dependencies</span><span class="chip">claude/whitepaper-implementation-plan-x8f7bh @ 26b76cd</span><span class="chip">live GitHub stats disabled</span></div></section><div class="grid" aria-label="Measured outcomes"><article class="cell"><div class="metric"><em>0.43 ms</em></div><strong>blast-radius lookup</strong><p class="muted">Measured from this repo's benchmark report, not a marketing placeholder.</p><p class="src">reports/benchmarks.md</p></article><article class="cell"><div class="metric"><em>118 ms</em></div><strong>pre-action gate</strong><p class="muted">Assumptions, routing, reuse, context, impact, scope, and anchoring.</p><p class="src">reports/benchmarks.md</p></article><article class="cell"><div class="metric"><em>62.1%</em></div><strong>cost saved vs always-premium</strong><p class="muted">Documented from the white-paper prototype and exposed by Forge cost reports.</p><p class="src">whitepaper prototype</p></article></div><section id="quickstart"><h2>Quickstart</h2><div class="terminal">npm install -g @codewithjuber/forgekit
+</style></head><body><div class="wrap"><header class="nav"><a class="brand" href="../">forge<em>kit</em></a><nav class="links" aria-label="Primary"><a href="../">Landing</a><a href="#quickstart">Quickstart</a><a href="#changes">Latest</a><a href="#sources">Data Sources</a><a href="https://github.com/CodeWithJuber/forgekit">GitHub ↗</a></nav></header><main id="top"><section class="hero" style="border-bottom:0;padding-bottom:0"><p class="eyebrow">@codewithjuber/forgekit · v0.8.0 · Node &gt;=20</p><h1>Live status, straight from the repository.</h1><p class="lead">One brain for every AI coding agent — the cognitive substrate every frozen model is missing (proof-carrying memory, impact foresight, enforced guardrails), authored once and delivered as native config to Claude Code, Codex, Cursor, Gemini, Aider, and more.</p><p><a class="btn primary" href="#quickstart">Install in 60 seconds</a> <a class="btn" href="https://github.com/CodeWithJuber/forgekit#readme">Read the docs</a></p><div class="meta"><span class="chip">MIT license</span><span class="chip">0 runtime dependencies</span><span class="chip">claude/forge-work-system-setup-p26ka5 @ 26586e2</span><span class="chip">live GitHub stats disabled</span></div></section><div class="grid" aria-label="Measured outcomes"><article class="cell"><div class="metric"><em>0.43 ms</em></div><strong>blast-radius lookup</strong><p class="muted">Measured from this repo's benchmark report, not a marketing placeholder.</p><p class="src">reports/benchmarks.md</p></article><article class="cell"><div class="metric"><em>118 ms</em></div><strong>pre-action gate</strong><p class="muted">Assumptions, routing, reuse, context, impact, scope, and anchoring.</p><p class="src">reports/benchmarks.md</p></article><article class="cell"><div class="metric"><em>62.1%</em></div><strong>cost saved</strong><p class="muted">Documented from the white-paper prototype and exposed by Forge cost reports.</p><p class="src">whitepaper prototype</p></article></div><section id="quickstart"><h2>Quickstart</h2><div class="terminal">npm install -g @codewithjuber/forgekit
 forge init
 forge doctor
-forge substrate "Change auth validation and update tests"</div></section><section id="changes"><h2>Latest repo changes</h2><div class="card"><ul class="list"><li>Optional embeddings tier (src/embed.js, ADR-0005; ROADMAP &quot;Next&quot;): set</li><li>forge uicheck visual &lt;file-or-url&gt; — the Playwright visual loop</li><li>Ledger read-path flip (P2). Reads are now a merged view (legacy ∪ ledger) via the</li></ul><p class="muted">Benchmark sections indexed: 3 · benchmarks file updated 2026-07-07.</p></div></section><section id="sources"><h2>Data Sources</h2><div class="card"><p class="muted">No mock data is used. This page is regenerated from repository files during CI (generated 2026-07-07T21:15:58.111Z from 26b76cd). Enable <code>BUILD_PAGES_LIVE=1</code> to refresh public GitHub counters with ETag/Last-Modified caching.</p><ul class="list"><li>package.json</li><li>README.md</li><li>CHANGELOG.md</li><li>reports/benchmarks.md</li><li>https://api.github.com/repos/CodeWithJuber/forgekit (optional, no auth, only when BUILD_PAGES_LIVE=1)</li></ul></div></section></main><footer>WCAG-minded semantic HTML, keyboard focus, responsive 320px–1920px+, and reduced-motion-safe. Same design tokens as the landing page — <code>forge uicheck design</code> gates both.</footer></div></body></html>
+forge substrate "Change auth validation and update tests"</div></section><section id="changes"><h2>Latest repo changes</h2><div class="card"><ul class="list"></ul><p class="muted">Benchmark sections indexed: 3 · benchmarks file updated 2026-07-08.</p></div></section><section id="sources"><h2>Data Sources</h2><div class="card"><p class="muted">No mock data is used. This page is regenerated from repository files during CI (generated 2026-07-08T18:42:41.820Z from 26586e2). Enable <code>BUILD_PAGES_LIVE=1</code> to refresh public GitHub counters with ETag/Last-Modified caching.</p><ul class="list"><li>package.json</li><li>README.md</li><li>CHANGELOG.md</li><li>reports/benchmarks.md</li><li>https://api.github.com/repos/CodeWithJuber/forgekit (optional, no auth, only when BUILD_PAGES_LIVE=1)</li></ul></div></section></main><footer>WCAG-minded semantic HTML, keyboard focus, responsive 320px–1920px+, and reduced-motion-safe. Same design tokens as the landing page — <code>forge uicheck design</code> gates both.</footer></div></body></html>

package/src/cli.js CHANGED Viewed

@@ -40,17 +40,40 @@ const COMMANDS = {
   brand: "print the active brand token map",
 };
+const GROUPS = {
+  Core: ["init", "sync", "doctor", "catalog"],
+  Memory: ["cortex", "recall", "remember", "brain", "ledger", "reuse"],
+  Substrate: [
+    "substrate",
+    "preflight",
+    "route",
+    "impact",
+    "scope",
+    "context",
+    "anchor",
+    "diagnose",
+    "imagine",
+    "lean",
+  ],
+  Quality: ["verify", "scan", "spec", "taste", "uicheck", "harden"],
+  Config: ["config", "cost", "dash", "brand", "atlas"],
+};
 const printVersion = () => console.log(`${BRAND.brand} (${BRAND.pkg}) v${BRAND.version}`);
 function printHelp() {
   printVersion();
   console.log(`\n${BRAND.tagline}\n`);
   console.log(`Usage: ${BRAND.cli} <command> [options]\n`);
-  console.log("Commands:");
-  for (const [name, desc] of Object.entries(COMMANDS)) {
-    console.log(`  ${name.padEnd(8)} ${desc}`);
+  for (const [group, cmds] of Object.entries(GROUPS)) {
+    console.log(`${group}:`);
+    for (const name of cmds) {
+      if (COMMANDS[name]) console.log(`  ${name.padEnd(12)} ${COMMANDS[name]}`);
+    }
+    console.log();
   }
-  console.log(`\nRun \`${BRAND.cli} <command> --help\` for details.`);
+  console.log(`Start here: \`${BRAND.cli} catalog\``);
+  console.log(`Run \`${BRAND.cli} <command> --help\` for details.`);
 }
 async function run(argv) {
@@ -139,7 +162,7 @@ async function run(argv) {
     console.log(`${BRAND.brand} sync — one source → every tool\n`);
     for (const r of report) {
       console.log(
-        `  ${r.action.padEnd(16)} ${String(r.target).padEnd(22)} ${r.tool}${r.note ? "  · " + r.note : ""}`,
+        `  ${r.action.padEnd(16)} ${String(r.target).padEnd(22)} ${r.tool}${r.note ? `  · ${r.note}` : ""}`,
       );
     }
     for (const w of warnings) console.warn(`  ! ${w}`);
@@ -158,7 +181,7 @@ async function run(argv) {
     const icon = { ok: "✓", warn: "!", fail: "✗" };
     console.log(`${BRAND.brand} doctor\n`);
     for (const r of results) console.log(`  ${icon[r.status]} ${r.label.padEnd(16)} ${r.note}`);
-    console.log(`\n${failed === 0 ? "all clear" : failed + " problem(s)"}`);
+    console.log(`\n${failed === 0 ? "all clear" : `${failed} problem(s)`}`);
     if (failed) process.exitCode = 1;
     return;
   }
@@ -596,7 +619,7 @@ async function run(argv) {
     if (r.findings?.length) {
       for (const f of r.findings) console.log(`  [${f.sev}] ${f.msg}`);
     } else if (r.raw) {
-      console.log("  " + r.raw.trim().split("\n").slice(-6).join("\n  "));
+      console.log(`  ${r.raw.trim().split("\n").slice(-6).join("\n  ")}`);
     } else {
       console.log("  no obvious red flags");
     }

package/src/cortex_mcp.js CHANGED Viewed

@@ -8,7 +8,7 @@ import { createInterface } from "node:readline";
 import { fileURLToPath } from "node:url";
 import { lessonsForContext, summary } from "./cortex.js";
 import { report as costReport } from "./cost_report.js";
-import { dashData } from "./dash.js";
+import { dashData, dashSummary } from "./dash.js";
 import { diagnose } from "./diagnose.js";
 import { doctor } from "./doctor.js";
 import { assessTask, clarifyBlock, preflightRepo } from "./preflight.js";
@@ -130,6 +130,12 @@ const TOOLS = [
       "Dashboard JSON payload — ledger stats, metrics, atlas info. Same data forge dash serves at /api/data.",
     inputSchema: { type: "object", properties: {} },
   },
+  {
+    name: "forge_dash_summary",
+    description:
+      "Lightweight dashboard health check — just counts (claims, tombstoned, contested, atlas built, metric events). Cheaper than forge_dash_data.",
+    inputSchema: { type: "object", properties: {} },
+  },
   {
     name: "forge_brain",
     description: "Project memory index — list all remembered facts stored in .forge/brain/.",
@@ -173,6 +179,44 @@ const TOOLS = [
       "Provider detection — which API provider is active (auto-detected or configured), env vars set, and health checks.",
     inputSchema: { type: "object", properties: {} },
   },
+  {
+    name: "forge_remember",
+    description:
+      "Store a durable fact in this repo's portable memory (.forge/brain/). Use for non-obvious, lasting knowledge (env quirks, decisions, gotchas).",
+    inputSchema: {
+      type: "object",
+      properties: {
+        name: { type: "string", description: "short slug for the fact (used as filename)" },
+        body: { type: "string", description: "the fact content (markdown)" },
+      },
+      required: ["name", "body"],
+    },
+  },
+  {
+    name: "forge_ledger_ratify",
+    description:
+      "Promote a ledger claim's confidence — record an independent oracle ratification (the claim held under test).",
+    inputSchema: {
+      type: "object",
+      properties: {
+        id: { type: "string", description: "claim ID or unique prefix" },
+      },
+      required: ["id"],
+    },
+  },
+  {
+    name: "forge_ledger_retract",
+    description:
+      "Tombstone a ledger claim with a reason — mark it as no longer valid so it stops influencing routing and memory.",
+    inputSchema: {
+      type: "object",
+      properties: {
+        id: { type: "string", description: "claim ID or unique prefix" },
+        reason: { type: "string", description: "why the claim is being retracted" },
+      },
+      required: ["id", "reason"],
+    },
+  },
 ];
 async function callTool(name, args = {}) {
@@ -212,6 +256,7 @@ async function callTool(name, args = {}) {
   }
   if (name === "forge_cost") return JSON.stringify(costReport(root), null, 2);
   if (name === "forge_dash_data") return JSON.stringify(dashData(root), null, 2);
+  if (name === "forge_dash_summary") return JSON.stringify(dashSummary(root), null, 2);
   if (name === "forge_brain") {
     try {
       const { brainStore, list, buildIndex } = await import("./brain.js");
@@ -287,6 +332,34 @@ async function callTool(name, args = {}) {
       2,
     );
   }
+  if (name === "forge_remember") {
+    const { brainStore, remember } = await import("./brain.js");
+    const store = brainStore(root);
+    remember(store, String(args.name ?? ""), String(args.body ?? ""));
+    return `Remembered "${args.name}" in ${store}.`;
+  }
+  if (name === "forge_ledger_ratify") {
+    const { ratify, repoLedger, getClaimByPrefix } = await import("./ledger_store.js");
+    const { gitAuthor } = await import("./util.js");
+    const dir = repoLedger(root);
+    const claim = getClaimByPrefix(dir, String(args.id ?? ""));
+    if (!claim) return `No claim matching prefix "${args.id}".`;
+    ratify(dir, claim.id, { author: gitAuthor(), t: today() });
+    return `Ratified claim ${claim.id}.`;
+  }
+  if (name === "forge_ledger_retract") {
+    const { tombstone, repoLedger, getClaimByPrefix } = await import("./ledger_store.js");
+    const { gitAuthor } = await import("./util.js");
+    const dir = repoLedger(root);
+    const claim = getClaimByPrefix(dir, String(args.id ?? ""));
+    if (!claim) return `No claim matching prefix "${args.id}".`;
+    tombstone(dir, claim.id, {
+      author: gitAuthor(),
+      reason: String(args.reason ?? ""),
+      t: today(),
+    });
+    return `Retracted claim ${claim.id}: ${args.reason}`;
+  }
   return null;
 }

package/src/dash.html CHANGED Viewed

@@ -210,7 +210,7 @@ const bar = (v, { dim = false } = {}) =>
 // Every claim row carries its provenance affordance — click to copy the exact command.
 const blameCell = (id8) =>
-  '<td><span class="blame" data-id8="' + esc(id8) + '">forge ledger blame ' + esc(id8) + "</span></td>";
+  `<td><span class="blame" data-id8="${esc(id8)}">forge ledger blame ${esc(id8)}</span></td>`;
 // The two writes (spec §2), per live row. Ratify = human-only hikma promotion (the
 // server stamps YOUR git identity as author); retract = tombstone with a reason.
@@ -240,37 +240,37 @@ let DATA = null;
 function renderLedger(d) {
   const L = d.ledger;
-  $("hdr-repo").textContent = "· " + d.repo;
+  $("hdr-repo").textContent = `· ${d.repo}`;
   $("hdr-meta").textContent =
-    L.stats.total + " claims · " + L.stats.tombstoned + " tombstoned · day " + d.nowDay;
+    `${L.stats.total} claims · ${L.stats.tombstoned} tombstoned · day ${d.nowDay}`;
   $("ledger-stats").innerHTML =
     '<span class="stat"><span class="n">' + L.stats.total + '</span><span class="l">claims</span></span>' +
     '<span class="stat"><span class="n ember">' + L.stats.val.trusted + '</span><span class="l">trusted ≥0.65</span></span>' +
     '<span class="stat"><span class="n">' + L.stats.val.uncertain + '</span><span class="l">uncertain</span></span>' +
     '<span class="stat"><span class="n">' + L.stats.val.dormant + '</span><span class="l">dormant &lt;0.35</span></span>' +
     Object.entries(L.stats.byKind).map(([k, n]) =>
-      '<span class="stat"><span class="n">' + n + '</span><span class="l">' + esc(k) + "</span></span>").join("");
+      `<span class="stat"><span class="n">${n}</span><span class="l">${esc(k)}</span></span>`).join("");
   const kinds = Object.keys(L.stats.byKind).sort();
   const sel = $("kind-filter");
   const keep = sel.value;
   sel.innerHTML = '<option value="">all</option>' +
-    kinds.map((k) => '<option value="' + esc(k) + '">' + esc(k) + "</option>").join("");
+    kinds.map((k) => `<option value="${esc(k)}">${esc(k)}</option>`).join("");
   if (kinds.includes(keep)) sel.value = keep;
   const shown = L.claims.filter((c) => !sel.value || c.kind === sel.value);
-  $("claim-count").textContent = shown.length + " of " + L.stats.total + " shown";
+  $("claim-count").textContent = `${shown.length} of ${L.stats.total} shown`;
   $("claims").innerHTML = shown.length ? claimTable(shown)
     : '<div class="empty">no claims yet — mint one: forge remember "&lt;name&gt;" "&lt;fact&gt;"</div>';
   $("contested").innerHTML = L.contested.length
-    ? '<span class="tag contested">' + L.contested.length + " contested</span>" + claimTable(L.contested)
+    ? `<span class="tag contested">${L.contested.length} contested</span>${claimTable(L.contested)}`
     : '<div class="empty">none — no live claim is both uncertain and contradicted</div>';
   const trust = Object.entries(L.trust).sort((a, b) => b[1] - a[1]);
   $("trust").innerHTML = trust.length
     ? "<table><tbody>" + trust.map(([a, u]) =>
-        '<tr><td class="muted">' + esc(a) + "</td><td>" + bar(u) + "</td></tr>").join("") +
+        `<tr><td class="muted">${esc(a)}</td><td>${bar(u)}</td></tr>`).join("") +
       "</tbody></table>"
     : '<div class="empty">no authored claims yet</div>';
 }
@@ -280,35 +280,42 @@ function renderCost(d) {
   const saved = stages.reduce((n, [, s]) => n + (s.savedEstimate || 0), 0);
   const events = stages.reduce((n, [, s]) => n + (s.events || 0), 0);
   const spend = d.spend;
-  $("cost-stats").innerHTML =
-    (spend && spend.totalCost > 0
-      ? '<span class="stat"><span class="n ember">$' + spend.totalCost.toFixed(2) + '</span><span class="l">real spend</span></span>' +
-        '<span class="stat"><span class="n">' + spend.sessions + '</span><span class="l">sessions</span></span>'
-      : '') +
-    '<span class="stat"><span class="n' + (spend && spend.totalCost > 0 ? '' : ' ember') + '">' + saved.toLocaleString() + '</span><span class="l">est. tokens saved</span></span>' +
-    '<span class="stat"><span class="n">' + events + '</span><span class="l">stage events</span></span>';
-  if (spend && spend.byModel && spend.byModel.length) {
+  const noData = (!spend || spend.totalCost <= 0) && saved === 0 && events === 0;
+  if (noData) {
+    $("cost-stats").innerHTML =
+      '<div class="empty">no cost data yet — run forge substrate or forge cost to start tracking</div>';
+  } else {
+    $("cost-stats").innerHTML =
+      (spend && spend.totalCost > 0
+        ? '<span class="stat"><span class="n ember">$' + spend.totalCost.toFixed(2) + '</span><span class="l">real spend</span></span>' +
+          '<span class="stat"><span class="n">' + spend.sessions + '</span><span class="l">sessions</span></span>'
+        : '') +
+      '<span class="stat"><span class="n' + (spend && spend.totalCost > 0 ? '' : ' ember') + '">' + saved.toLocaleString() + '</span><span class="l">est. tokens saved</span></span>' +
+      '<span class="stat"><span class="n">' + events + '</span><span class="l">stage events</span></span>';
+  }
+  if (spend?.byModel?.length) {
     $("spend-section").innerHTML = '<h3>Spend by model</h3>' +
       '<table><thead><tr><th>model</th><th>cost</th><th>in tok</th><th>out tok</th></tr></thead><tbody>' +
       spend.byModel.map((m) => '<tr><td class="mono">' + esc(m.model) + '</td><td class="num">$' + m.cost.toFixed(4) +
           '</td><td class="num">' + (m.inTokens || 0).toLocaleString() +
           '</td><td class="num">' + (m.outTokens || 0).toLocaleString() + '</td></tr>').join('') + '</tbody></table>';
   } else {
-    $("spend-section").innerHTML = '';
+    $("spend-section").innerHTML =
+      '<div class="empty">no model spend data yet — real costs appear when ccusage is configured</div>';
   }
   $("stages").innerHTML = stages.length
     ? "<table><thead><tr><th>stage</th><th>events</th><th>outcomes</th><th>saved</th></tr></thead><tbody>" +
       stages.map(([name, s]) =>
         '<tr><td class="mono">' + esc(name) + '</td><td class="num">' + s.events + "</td><td>" +
         Object.entries(s.byOutcome || {}).map(([o, n]) =>
-          '<span class="tag">' + esc(o) + " " + n + "</span>").join(" ") +
+          `<span class="tag">${esc(o)} ${n}</span>`).join(" ") +
         '</td><td class="num">' + (s.savedEstimate || 0).toLocaleString() + "</td></tr>").join("") +
       "</tbody></table>"
     : '<div class="empty">no metrics yet — stages append to .forge/metrics.jsonl as you work</div>';
   $("recent").innerHTML = d.metrics.recent.slice().reverse().map((e) =>
     "<li>" + new Date(e.t).toLocaleTimeString() + ' <span class="st">' + esc(e.stage) + "</span> " +
-    esc(e.outcome || "") + (e.savedEstimate ? ' <span class="plus">+' + e.savedEstimate + "</span>" : "") +
-    "</li>").join("") || '<li class="empty">—</li>';
+    esc(e.outcome || "") + (e.savedEstimate ? ` <span class="plus">+${e.savedEstimate}</span>` : "") +
+    "</li>").join("") || '<li class="empty">no recent events — the substrate records cache hits, gate passes, and routing decisions as you work</li>';
 }
 async function runImpact() {
@@ -320,9 +327,9 @@ async function runImpact() {
     return;
   }
   out.innerHTML = '<div class="empty">tracing…</div>';
-  const res = await fetch("/api/impact?target=" + encodeURIComponent(target));
+  const res = await fetch(`/api/impact?target=${encodeURIComponent(target)}`);
   const r = await res.json();
-  if (!res.ok) { out.innerHTML = '<div class="empty">' + esc(r.error || "error") + "</div>"; return; }
+  if (!res.ok) { out.innerHTML = `<div class="empty">${esc(r.error || "error")}</div>`; return; }
   if (!r.found) {
     out.innerHTML = '<div class="empty">"' + esc(target) + '" not found in the atlas (' +
       r.totalGraphNodes + " nodes)</div>";
@@ -338,7 +345,7 @@ async function runImpact() {
       '</td><td class="mono">' + esc(x.node.name || x.id) + '</td><td class="muted">' +
       esc(x.node.file || "") + "</td></tr>").join("") +
     "</tbody></table>" +
-    (r.impacted.length > 40 ? '<div class="empty">… ' + (r.impacted.length - 40) + " more</div>" : "") +
+    (r.impacted.length > 40 ? `<div class="empty">… ${r.impacted.length - 40} more</div>` : "") +
     '<div class="empty" style="margin-top:8px">same trace on the CLI: forge impact ' + esc(target) + "</div>";
 }
@@ -376,7 +383,7 @@ document.addEventListener("click", (ev) => {
   }
   const el = ev.target.closest(".blame");
   if (!el) return;
-  navigator.clipboard?.writeText("forge ledger blame " + el.dataset.id8).then(() => {
+  navigator.clipboard?.writeText(`forge ledger blame ${el.dataset.id8}`).then(() => {
     el.classList.add("copied");
     const prev = el.textContent;
     el.textContent = "copied ✓";

package/src/dash.js CHANGED Viewed

@@ -111,6 +111,42 @@ export function dashData(root, { nowDay = epochDay() } = {}) {
   return { repo: basename(root), nowDay, ledger, metrics, atlas, spend };
 }
+/**
+ * Lightweight summary — counts and status only (no full claim list).
+ * @param {string} root
+ * @param {{nowDay?: number}} [opts]
+ */
+export function dashSummary(root, { nowDay = epochDay() } = {}) {
+  let ledgerStats = emptyLedger().stats;
+  let contested = 0;
+  try {
+    const dir = repoLedger(root);
+    ledgerStats = stats(dir, nowDay);
+    const all = loadClaims(dir);
+    contested = all.filter((c) => {
+      if (c.tombstone) return false;
+      const v = val(c, nowDay);
+      if (v < CONTESTED_BAND[0] || v > CONTESTED_BAND[1]) return false;
+      return (c.evidence ?? []).some((e) => validOutcome(e) && e.result === "contradict");
+    }).length;
+  } catch {}
+  let atlasBuilt = false;
+  try {
+    atlasBuilt = Boolean(loadAtlas(root));
+  } catch {}
+  let metricEvents = 0;
+  try {
+    metricEvents = readMetrics(root).length;
+  } catch {}
+  return {
+    claims: ledgerStats.total,
+    tombstoned: ledgerStats.tombstoned,
+    contested,
+    atlasBuilt,
+    metricEvents,
+  };
+}
 const HTML_PATH = join(dirname(fileURLToPath(import.meta.url)), "dash.html");
 const sendJson = (res, code, body) => {

package/src/emit/copilot.js CHANGED Viewed

@@ -1,7 +1,7 @@
 // GitHub Copilot's coding agent reads the root AGENTS.md natively (since 2025-08-28).
 export default {
   tool: "Copilot",
-  emit(ctx) {
+  emit(_ctx) {
     return {
       tool: this.tool,
       target: "AGENTS.md",

package/src/emit/gemini.js CHANGED Viewed

@@ -29,7 +29,7 @@ export default {
     const had = current.includes("AGENTS.md");
     context.fileName = [...new Set([...current, "AGENTS.md"])];
     settings._forge = ctx.shared.markerString(ctx.hash);
-    const action = ctx.shared.writeIfChanged(path, JSON.stringify(settings, null, 2) + "\n");
+    const action = ctx.shared.writeIfChanged(path, `${JSON.stringify(settings, null, 2)}\n`);
     return {
       tool: this.tool,
       target: ".gemini/settings.json",

package/src/emit/mcp.js CHANGED Viewed

@@ -33,7 +33,7 @@ function mergeJson(path, key, servers) {
   }
   if (!added) return { action: "unchanged", note: "present" };
   mkdirSync(dirname(path), { recursive: true });
-  writeFileSync(path, JSON.stringify(obj, null, 2) + "\n");
+  writeFileSync(path, `${JSON.stringify(obj, null, 2)}\n`);
   return { action: "written", note: `+${added} server(s)` };
 }
@@ -58,7 +58,7 @@ function emitContinueYaml(dir, servers) {
     lines.push(`  - name: ${name}`, "    type: stdio", `    command: ${def.command}`, "    args:");
     for (const a of def.args || []) lines.push(`      - ${JSON.stringify(a)}`);
   }
-  const content = lines.join("\n") + "\n";
+  const content = `${lines.join("\n")}\n`;
   if (existsSync(path) && readFileSync(path, "utf8") === content)
     return { action: "unchanged", note: "present" };
   mkdirSync(dir, { recursive: true });

package/src/init.js CHANGED Viewed

@@ -57,17 +57,28 @@ function unionStrings(a = [], b = []) {
   return [...set];
 }
-/** Merge Forge hook entries into existing hook arrays, matching by command to avoid duplicates. */
+/** Extract guard identity (basename + trailing args) from a hook command for dedup.
+ *  `bash ~/.forge/guards/cortex.sh prompt` and
+ *  `"${CLAUDE_PLUGIN_ROOT}"/global/guards/cortex.sh prompt` both → `cortex.sh prompt`. */
+function guardKey(command) {
+  const m = command.match(/([^/\\"]+\.sh)\s*(.*)/);
+  return m ? `${m[1]} ${m[2]}`.trim() : command;
+}
+/** Merge Forge hook entries into existing hook arrays, matching by guard identity to avoid duplicates. */
 function mergeHooks(existing = {}, template = {}) {
   const merged = { ...existing };
   for (const [event, entries] of Object.entries(template)) {
     const existingEntries = merged[event] || [];
-    const existingCommands = new Set(
-      existingEntries.flatMap((e) => (e.hooks || []).map((h) => h.command)).filter(Boolean),
+    const existingKeys = new Set(
+      existingEntries
+        .flatMap((e) => (e.hooks || []).map((h) => h.command))
+        .filter(Boolean)
+        .map(guardKey),
     );
     const newEntries = [];
     for (const entry of entries) {
-      const hooks = (entry.hooks || []).filter((h) => !existingCommands.has(h.command));
+      const hooks = (entry.hooks || []).filter((h) => !existingKeys.has(guardKey(h.command)));
       if (hooks.length) {
         newEntries.push({ ...entry, hooks });
       }

package/src/sync.js CHANGED Viewed

@@ -31,7 +31,7 @@ export function assemble(rules) {
     for (const rule of section.rules || []) out.push(`- ${rule}`);
     out.push("");
   }
-  return out.join("\n").trimEnd() + "\n";
+  return `${out.join("\n").trimEnd()}\n`;
 }
 function loadRules(targetRoot) {

package/templates/project-layer/CLAUDE.md CHANGED Viewed

@@ -1,27 +1,24 @@
-# Hostlelo — project instructions
+# <Project> — project instructions
-Drop this into a Hostlelo repo root. Adds project context on top of the global
+Drop this into a repo root. Adds project context on top of the global
 `~/.claude/CLAUDE.md`. Keep it lean; prune anything Claude gets right without it.
 Shared cross-tool rules (stack, commands, engineering rules) live in **@AGENTS.md**
 so Cursor/Codex/Gemini read the same source. This file holds only Claude-specific
-+ Hostlelo-specific bits. Keep CLAUDE.md thin; put general rules in AGENTS.md.
++ project-specific bits. Keep CLAUDE.md thin; put general rules in AGENTS.md.
 ## Stack
-- Runtime: Node.js. DB: self-hosted Postgres (18+) with **pgvector** on the VPS.
-- LLM: routed through **OpenRouter** — never hardcode a single provider; use the
-  configured routing/env, and pick cheap models for classify/embed, stronger
-  models only for hard generation.
-- Delivery surfaces: customer AI **chat widget** (`widget.js`) and **Telegram**
-  owner control plane. Infra: Linode VPS + Cloudflare Tunnel/Access.
+- Runtime: <runtime>. DB: <database>.
+- LLM: routed through <provider> — never hardcode a single provider; use the
+  configured routing/env.
+- Delivery surfaces: <describe your app surfaces>.
 ## Non-negotiables
-- **Grounding-first.** Customer-facing answers (esp. prices/plans) must come from
-  the verified KB, never guessed. If unverified, say so or defer — do not invent
-  numbers, features, or availability.
+- **Grounding-first.** Customer-facing answers must come from verified data, never
+  guessed. If unverified, say so or defer — do not invent numbers, features, or
+  availability.
 - **Secrets** live in env / the secrets store, never in code, logs, or commits.
   `.env*` and key files are hook-protected — don't work around it.
-- Neutral, worldwide tone in customer replies (no forced regional greeting).
 ## Workflow
 - Prefer readymade, well-maintained OSS libraries/SDKs over hand-rolled code —
@@ -29,7 +26,7 @@ so Cursor/Codex/Gemini read the same source. This file holds only Claude-specifi
 - Before DB or deploy changes: read the migration/deploy path first; never run
   destructive SQL (`DROP`/`TRUNCATE`) or touch prod without explicit confirmation.
 - Verify every change: run the app's tests/typecheck/build and show output.
-- Deploy via the `hostlelo-deploy` skill (safe, staged, with rollback notes).
+- Deploy via the `<your deploy skill>` skill (safe, staged, with rollback notes).
 ## Commands
 <!-- Fill these in from package.json so Claude doesn't guess: -->

package/skills/cognitive-substrate/SKILL.md DELETED Viewed

@@ -1,56 +0,0 @@
----
-name: cognitive-substrate
-description: >-
-  Use before ambiguous, expensive, multi-file, or mutating coding work to run
-  Forge's cognitive substrate: assumption gate, model routing, impact
-  prediction, scope decomposition, memory/lesson lookup, minimality check, and
-  verification planning. Trigger when a user asks to edit/refactor/fix/design
-  production code, integrate features, choose model effort, inspect blast
-  radius, or avoid agent assumptions.
----
-# Cognitive Substrate
-Wrap the frozen model in a pre-action check: gate assumptions, route model effort, predict
-blast radius, decompose scope, surface past lessons, and plan verification — **before** editing.
-## When to run it
-Any ambiguous, expensive, multi-file, or mutating task. Skip one-line, well-specified fixes.
-In Claude Code it fires automatically on every prompt (UserPromptSubmit hook); in other
-agents run the CLI yourself, or call the MCP tool `substrate_check`.
-## Run
-```bash
-forge substrate "<task>" --json     # full contract; use the fields below
-```
-## Act on the result
-1. **`okToProceed: false`** → ask the returned `assumption.questions` before editing. Do not guess.
-2. **`route.tier`** → start at that model tier (cheapest capable); escalate only after an external verifier fails.
-3. **`impact.impactedFiles`** → read these before editing a named symbol/file (the blast radius).
-4. **`scope.clusters`** → split independent groups into separate sessions; note coupled files you didn't name.
-5. **`memory.advisory`** → context, not law; tests and human corrections override it.
-6. **`verification.checklist`** → run it and show output before claiming done.
-## Worked example
-`forge substrate "make the auth better"` → `proceed: ASK FIRST` + clarify
-questions: ask them instead of editing. A clear task returns `proceed: yes` plus
-the impacted files — including importers you didn't name.
-## Deeper single checks
-`forge preflight "<task>"` (assumptions) · `forge route "<task>"` (model tier) ·
-`forge impact <symbol|file>` (blast radius) · `forge scope <file…>` (decomposition) ·
-`forge context "<task>"` (budgeted context assembly; *computes* what's missing) ·
-`forge imagine "<task>" [--run]` (predicted breaks + minimal covering test suite;
-`--run` dry-runs it in a sandboxed worktree) ·
-`forge diagnose "<error>"` (doom-loop: 3× the same failure signature = stop retrying,
-escalate one tier with the minted diagnosis claim at the head of the prompt).
-MCP equivalents: `assumption_gate`, `route_task`, `predict_impact`, `scope_files`.
-For the full guide (how it works, extending it, the honesty boundary) and the white paper,
-see `docs/cognitive-substrate/README.md`. `references/capability-map.md` maps faculties to commands.

package/skills/cognitive-substrate/references/capability-map.md DELETED Viewed

@@ -1,17 +0,0 @@
-# Cognitive substrate capability map
-| Paper capability | Forge surface | Guarantee |
-| --- | --- | --- |
-| Memory | `forge recall`, `forge cortex`, `forge ledger` | Facts/lessons persist as content-addressed ledger claims; `forge ledger blame` shows provenance. |
-| Learning | `forge cortex`, ledger oracles | External outcomes (tests, CI, human accept/revert) move claim confidence; model weights do not change. |
-| Imagination | `forge imagine [--run]`, `forge impact` | Predicted breaks + minimal covering test suite; `--run` dry-runs it in a sandboxed worktree. |
-| Self-correction | `forge verify`, `forge diagnose` | Tests/builds beat model claims; 3× the same failure signature mints a diagnosis + escalation. |
-| Impact-awareness | `forge atlas`, `forge impact` | Known symbols/files and likely dependents are surfaced. |
-| M1 routing | `forge route` | Transparent model-tier recommendation. |
-| M2 assumption gate | `forge preflight`, `forge context` | Under-specified tasks return *computed* missing-set questions. |
-| M3 decomposition | `forge scope` | Import clusters show independent vs coupled files. |
-| M4 goal anchoring | `forge anchor`, `forge substrate` | Changed files are checked against the stated goal. |
-| M5 anti-over-engineering | `forge lean`, `forge uicheck design` | Footprint vs ask; UI slop-distance + fingerprint conformance gate. |
-| M6 inline verification | `forge verify` | External checks are required before done. |
-Limits: static graph edges are conservative; memory relevance and model routing are advisory; non-hook tools cannot be forcibly blocked.