anymorph 0.4.0 → 0.5.0

package/dist/skillpacks/geo/skills/geo-initializing-strategy/references/semantic-clusters-diagnosis.md

@@ -1,58 +0,0 @@
-# Strategy Map: Semantic Clusters Diagnosis
-
-Map the brand's long-term opportunity space into semantic clusters.
-
-Each cluster should describe:
-
-- intent
-- AI decision criteria
-- current coverage state
-- missing surface types
-- known overlap or cannibalization risk
-- confidence
-
-Do not include recommended actions, briefs, or page creation instructions. The orchestrator will turn cluster diagnosis into long-term strategy.
-
-Use DataForSEO-backed tools first for keyword demand, SERP shape, ranked-keyword gaps, and competitor content gaps:
-
-- `mcp__anymorph__search_keywords`
-- `mcp__anymorph__serp_snapshot`
-- `mcp__anymorph__dataforseo_research`
-- `mcp__anymorph__get_competitor_gap`
-
-Use Ahrefs only when available for in-content internal-link hub evidence. If Ahrefs returns "API units limit reached", times out, fails auth, or is unavailable, stop calling Ahrefs and omit `hubSurfaceId` unless a real non-Ahrefs surface id is known.
-
-Never emit `null`. Omit optional fields like `hubSurfaceId` and `overlaps` when unknown. Required unknown enum fields must use `unknown`.
-
-Submit exactly one section artifact with `mcp__strategy_map_artifacts__submit_strategy_map_section`.
-
-```json
-{
-  "section": "semanticClusters",
-  "artifact": [
-    {
-      "id": "cl-short-stable-id",
-      "name": "Cluster name",
-      "intent": "User intent this cluster captures",
-      "aiDecisionCriteria": ["Criteria AI answers compare"],
-      "priority": "high | medium | low",
-      "coverageState": "covered | gap | weak | overlap | monitor | unknown",
-      "hubSurfaceId": "optional existing hub surface id",
-      "missingSurfaceTypes": ["proof_surface", "comparison_page"],
-      "overlaps": ["optional overlapping cluster ids"],
-      "confidence": "high | medium | low",
-      "evidenceRefs": ["ev-semantic-1"]
-    }
-  ],
-  "evidence": [
-    {
-      "id": "ev-semantic-1",
-      "type": "signal | workspace | source | research",
-      "ref": "tool or source reference",
-      "summary": "Specific evidence summary."
-    }
-  ]
-}
-```
-
-Final response: one short sentence naming the saved section.

package/dist/skillpacks/geo/skills/geo-initializing-strategy/references/strategy-map-contract.md

@@ -1,26 +0,0 @@
-# Strategy Map Contract
-
-The Strategy Map is a diagnosis artifact, not an execution plan.
-
-Include:
-
-- Owned-site foundation status and blockers.
-- Current AI visibility posture.
-- Semantic clusters with priority and coverage state.
-- External authority gaps.
-- Long-term investment mix and guardrails.
-
-Do not include:
-
-- Recommended page actions.
-- Outreach copy.
-- Concrete prompt panels.
-- Measurement targets.
-- Weekly success criteria.
-
-Evidence rules:
-
-- Use DataForSEO On-Page as the primary owned-site crawl and internal-link source.
-- Use Ahrefs only as optional enrichment for in-content internal-link and crawl evidence.
-- Use workspace MCP tools before freeform research.
-- Mark missing evidence as `unknown`; do not infer certainty from absence.

package/dist/skillpacks/geo/skills/geo-initializing-strategy/references/visibility-diagnosis.md

@@ -1,50 +0,0 @@
-# Strategy Map: Visibility Diagnosis
-
-Summarize current AI visibility.
-
-Classify durable patterns by cluster or engine:
-
-- `absent`
-- `cited_not_mentioned`
-- `mentioned_not_recommended`
-- `cited_not_recommended`
-- `recommended`
-- `declining`
-- `unknown`
-
-Do not paste raw AI answers. Do not propose weekly actions.
-
-Never emit `null`. Omit optional fields like `clusterId` and `engine` when unknown. Required unknown enum fields must use `unknown`.
-
-Submit exactly one section artifact with `mcp__strategy_map_artifacts__submit_strategy_map_section`.
-
-```json
-{
-  "section": "visibility",
-  "artifact": {
-    "overall": "strong | medium | weak | unknown",
-    "patterns": [
-      {
-        "id": "vis-short-stable-id",
-        "clusterId": "optional-cluster-id",
-        "engine": "optional engine name",
-        "state": "absent | cited_not_mentioned | mentioned_not_recommended | cited_not_recommended | recommended | declining | unknown",
-        "summary": "Cluster- or engine-level visibility pattern.",
-        "competitors": ["Competitor names"],
-        "evidenceRefs": ["ev-visibility-1"]
-      }
-    ],
-    "evidenceRefs": ["ev-visibility-1"]
-  },
-  "evidence": [
-    {
-      "id": "ev-visibility-1",
-      "type": "signal | workspace | source | research",
-      "ref": "tool or source reference",
-      "summary": "Specific evidence summary."
-    }
-  ]
-}
-```
-
-Final response: one short sentence naming the saved section.

package/dist/skillpacks/geo/skills/geo-local-setup/SKILL.md

@@ -1,66 +0,0 @@
----
-name: geo-local-setup
-description: Use when setting up, checking, repairing, or preparing a tenant repo for local Anymorph GEO work.
----
-
-# GEO Local Setup
-
-Use this skill before local GEO action generation in a tenant repo.
-
-## CLI
-
-Use the top-level `anymorph` CLI. Run commands from the tenant repo root.
-
-## Setup
-
-Run these from the tenant repo root:
-
-```bash
-anymorph status
-anymorph login
-anymorph init
-anymorph doctor
-```
-
-If credentials are missing, authenticated commands start device login
-automatically. Use `login` directly when you want to authenticate first.
-For non-interactive runs, set `ANYMORPH_ACCESS_TOKEN` and optionally
-`ANYMORPH_API_URL`.
-
-`init` installs managed skills, contracts, run folders, and missing memory files.
-`doctor` reports exact missing, extra, or drifted managed files. Use
-`doctor --fix` to repair managed files without touching memory or run artifacts.
-
-To compare against a known source checkout instead of the installed copy:
-
-```bash
-anymorph doctor --skills-source /path/to/anymorph-geo-skills/skills
-```
-
-## Prepare
-
-```bash
-anymorph status
-anymorph workspaces
-anymorph prepare <workspaceId>
-```
-
-After the agent writes `actions.json`:
-
-```bash
-anymorph validate {runId}
-git add agent/runs/{runId} agent/STRATEGY.md agent/LEARNINGS.md
-git commit -m "chore: add geo strategy run {runId}"
-git push origin main
-anymorph submit {runId}
-```
-
-`submit` registers or updates the backend run and immediately materializes it
-in Dashboard. There is no separate approve step.
-A submitted run is immutable: once it is `executed`, re-submitting the same
-`runId` (even with an edited `actions.json`) is a no-op. To add or change
-actions, run a fresh `prepare` for a new `runId` and submit that new run.
-For commerce workspaces, `validate` requires every `geo_page` create action to
-include `target.targetProductIds` from the prepared product files.
-
-The `anymorph` CLI owns local GEO setup, prepare, validate, and submit.

package/dist/skillpacks/geo/skills/geo-local-setup/agents/openai.yaml

@@ -1,4 +0,0 @@
-interface:
-  display_name: "GEO Local Setup"
-  short_description: "Set up and verify a tenant repo for local GEO work."
-  default_prompt: "Use $geo-local-setup to set up or check this tenant repo for local GEO work."

package/dist/skillpacks/geo/skills/geo-page-writer/SKILL.md

@@ -1,81 +0,0 @@
----
-name: geo-page-writer
-description: Use when writing or updating the actual MDX for an Anymorph-managed GEO page in a CMS tenant repo after a geo_page action or generated-page opportunity has been selected.
----
-
-# GEO Page Writer
-
-Use this skill to write or revise a real CMS MDX page for a selected
-`geo_page` action. This is an execution skill, not a strategy-planning skill.
-
-## Scope
-
-Write or update:
-
-- A tenant repo MDX page for a generated GEO page.
-- The supporting page-writer evidence artifacts under `agent/page-writer/`.
-
-Do not:
-
-- Write `actions.json`.
-- Submit GEO strategy runs.
-- Replace the backend CMS generation pipeline.
-- Invent claims, product facts, prices, image URLs, testimonials, quotes, or
-  source names.
-
-## References
-
-Read these references before writing:
-
-- `references/research.md` for source collection, Exa Deep Search, KB search,
-  image search, and source-pack rules.
-- `references/writing.md` for MDX structure, frontmatter, components, images,
-  CTA, and GEO writing rules.
-- `references/validation.md` for agent-owned brand and fact validation plus
-  script-backed MDX static checks.
-
-## Workflow
-
-1. Identify the selected `geo_page` action from `agent/runs/{runId}/actions.json`.
-2. Collect page sources:
-
-```bash
-node .agents/skills/geo-page-writer/scripts/collect-page-sources.mjs all \
-  --run-id <runId> \
-  --action-id <actionId> \
-  --workspace <workspace-id-or-domain>
-```
-
-Use `.claude/skills/...` instead of `.agents/skills/...` when running under
-Claude Code.
-
-3. Read `agent/page-writer/{runId}/{actionId}/sources/source-pack.json`.
-4. Write or edit the MDX manually using the source pack and `references/writing.md`.
-5. Self-review brand and fact claims with `references/validation.md`.
-6. Run the static MDX checker:
-
-```bash
-node .agents/skills/geo-page-writer/scripts/check-page-mdx.mjs \
-  --file <path/to/page.mdx> \
-  --sources agent/page-writer/<runId>/<actionId>/sources/source-pack.json
-```
-
-7. Fix every error. Review warnings before commit.
-
-## Artifact Contract
-
-Write page-writer artifacts here:
-
-```text
-agent/page-writer/{runId}/{actionId}/
-  sources/
-    exa-research.md
-    exa-response.json
-    kb-results.json
-    image-results.json
-    source-pack.json
-  checks/
-    mdx-static.json
-```
-
-Commit the MDX and relevant `agent/page-writer/...` artifacts together.

package/dist/skillpacks/geo/skills/geo-page-writer/agents/openai.yaml

@@ -1,4 +0,0 @@
-interface:
-  display_name: "GEO Page Writer"
-  short_description: "Write and validate real CMS MDX for selected GEO page actions."
-  default_prompt: "Use $geo-page-writer to collect sources, write the GEO page MDX, and run static MDX validation."

package/dist/skillpacks/geo/skills/geo-page-writer/references/research.md

@@ -1,61 +0,0 @@
-# Research
-
-The research phase builds the source pack used to write the page. It does not
-write the MDX.
-
-## Source Order
-
-Use sources in this order:
-
-1. Selected `geo_page` action and run context.
-2. Tenant workspace KB / living document.
-3. Product catalog or action `targetProductIds`.
-4. Exa Deep Search for external evidence.
-5. Image search for existing brand assets.
-
-Prefer first-party and workspace facts over external summaries. Use external
-research to fill market, comparison, definition, and citation gaps.
-
-## Exa Deep Search
-
-`collect-page-sources.mjs exa` uses Exa `/search` with `type:
-"deep-reasoning"`. Exa's older `/research/v1` API is deprecated, so do not use
-it for new work.
-
-The Exa result is a research input, not page copy. Treat its synthesized answer
-as a draft finding that still needs source-aware judgment.
-
-## KB Search
-
-Use KB search for brand-specific facts:
-
-- product positioning
-- process and methodology
-- pricing or packaging claims
-- company history
-- differentiators
-- support, FAQ, and policy details
-
-If KB search is unavailable, mark the missing query in the source pack and avoid
-brand-specific factual claims that need that evidence.
-
-## Image Search
-
-Use image search only for existing workspace assets. Do not generate images in
-this skill.
-
-Every MDX image URL should come from:
-
-- the source pack image results
-- product catalog image inputs
-- an existing tenant asset already present in the MDX/repo
-
-## Source Pack Rules
-
-The source pack must make writing decisions auditable:
-
-- Keep source URLs and titles.
-- Keep KB snippets separate from external web sources.
-- Keep image results separate from factual sources.
-- Record failures instead of silently omitting a source class.
-- Do not promote a source into a claim unless the source actually supports it.

package/dist/skillpacks/geo/skills/geo-page-writer/references/validation.md

@@ -1,59 +0,0 @@
-# Validation
-
-Validation has two owners:
-
-- The agent owns brand and fact validation.
-- `check-page-mdx.mjs` owns deterministic MDX static validation.
-
-## Brand Validation
-
-Before running the script, review the MDX against the source pack and tenant
-brand context.
-
-Check:
-
-- Tone matches the brand guidance.
-- Blocklisted phrases are absent.
-- Required positioning is represented.
-- "Must avoid" claims or angles are not used.
-- The brand role is honest and not forced into unrelated claims.
-- Competitive comparisons are measured and sourced.
-
-If unsure, soften the claim or remove it.
-
-## Fact Validation
-
-Every concrete factual claim must be supported by the source pack or existing
-tenant/product data.
-
-Remove or soften:
-
-- numbers not present in sources
-- dates not present in sources
-- "best", "#1", "first", "only", "guaranteed", or "clinically proven" without
-  direct evidence
-- product efficacy claims without product/KB/source support
-- competitor comparisons without a source for both sides
-- exact quotes not present verbatim in a source
-
-Use hedged language when evidence is directional but not conclusive.
-
-## MDX Static Validation
-
-Run:
-
-```bash
-node .agents/skills/geo-page-writer/scripts/check-page-mdx.mjs \
-  --file <path/to/page.mdx> \
-  --sources agent/page-writer/<runId>/<actionId>/sources/source-pack.json
-```
-
-The checker validates:
-
-- frontmatter presence
-- MDX/JSX parse when a compiler is available
-- import and component allowlists
-- image URL provenance
-- link and CTA URL shape
-
-Fix all errors. Review warnings before committing.

package/dist/skillpacks/geo/skills/geo-page-writer/references/writing.md

@@ -1,55 +0,0 @@
-# Writing
-
-Write the actual MDX directly. Do not use a writing script.
-
-## Page Shape
-
-The page should be useful to a human reader and easy for AI engines to cite.
-
-Use:
-
-- Answer-first opening.
-- Clear H2 sections that each answer one question.
-- Self-contained paragraphs with explicit entities.
-- Specific product, process, or methodology details when sourced.
-- Tables or lists only when they improve comparison or extraction.
-
-Avoid:
-
-- Generic SEO filler.
-- Keyword stuffing.
-- Unsupported rankings or superlatives.
-- Fake quotes, fake studies, fake customers, or invented numbers.
-- Images that are merely decorative when product/source evidence is needed.
-
-## Frontmatter
-
-Include the tenant's expected MDX frontmatter fields. At minimum:
-
-- `title`
-- `description`
-- `lang`
-
-Include `heroImage` only when the URL is from the source pack, product catalog,
-or an existing tenant asset.
-
-## Components
-
-Use only components already supported by the tenant CMS renderer. Prefer plain
-MDX when unsure.
-
-For commerce/product pages:
-
-- Only mention products present in the source pack or product context.
-- Use product image, price, URL, and title verbatim.
-- Do not compute discounts or localize prices yourself.
-
-## CTA
-
-Use CTA URLs from tenant inputs or existing safe routes. Do not invent checkout,
-demo, pricing, or contact URLs.
-
-## Images
-
-Use image URLs verbatim. Add useful alt text. Do not crop product imagery in a
-way that hides packaging, texture, label, or use context.

package/dist/skillpacks/geo/skills/geo-page-writer/scripts/check-page-mdx.mjs

@@ -1,210 +0,0 @@
-#!/usr/bin/env node
-import { mkdir, readFile, writeFile } from "node:fs/promises";
-import { dirname, isAbsolute, join, resolve } from "node:path";
-import { pathToFileURL } from "node:url";
-
-const args = parseArgs(process.argv.slice(2));
-
-try {
-	const result = await check(args);
-	console.log(JSON.stringify(result, null, 2));
-	process.exit(result.ok ? 0 : 1);
-} catch (err) {
-	const message = err instanceof Error ? err.message : String(err);
-	console.error(message);
-	process.exit(1);
-}
-
-function parseArgs(argv) {
-	const out = {};
-	for (let i = 0; i < argv.length; i += 1) {
-		const arg = argv[i];
-		if (arg === "--file") out.file = requireValue(argv, ++i, arg);
-		else if (arg === "--sources") out.sources = requireValue(argv, ++i, arg);
-		else if (arg === "--out") out.out = requireValue(argv, ++i, arg);
-		else throw new Error(`Unknown argument: ${arg}`);
-	}
-	return out;
-}
-
-function requireValue(argv, index, flag) {
-	const value = argv[index];
-	if (!value || value.startsWith("--")) throw new Error(`${flag} requires a value`);
-	return value;
-}
-
-async function check(input) {
-	const file = required(input.file, "--file");
-	const sourcePath = input.sources;
-	const mdxPath = resolve(file);
-	const mdx = await readFile(mdxPath, "utf8");
-	const sourcePack = sourcePath ? JSON.parse(await readFile(resolve(sourcePath), "utf8")) : null;
-	const errors = [];
-	const warnings = [];
-	const frontmatter = parseFrontmatter(mdx, errors);
-	checkFrontmatter(frontmatter, errors);
-	checkImports(mdx, errors, warnings);
-	checkComponents(mdx, errors);
-	checkLinks(mdx, warnings);
-	checkImages(mdx, sourcePack, warnings);
-	await checkMdxCompile(mdx, errors, warnings);
-	const result = {
-		ok: errors.length === 0,
-		file: mdxPath,
-		checkedAt: new Date().toISOString(),
-		errors,
-		warnings,
-	};
-	if (input.out) {
-		const outPath = isAbsolute(input.out) ? input.out : resolve(input.out);
-		await mkdir(dirname(outPath), { recursive: true });
-		await writeFile(outPath, `${JSON.stringify(result, null, 2)}\n`);
-	}
-	return result;
-}
-
-function parseFrontmatter(mdx, errors) {
-	if (!mdx.startsWith("---\n")) {
-		errors.push("frontmatter: file must start with YAML frontmatter");
-		return {};
-	}
-	const end = mdx.indexOf("\n---", 4);
-	if (end === -1) {
-		errors.push("frontmatter: closing --- not found");
-		return {};
-	}
-	const raw = mdx.slice(4, end).trim();
-	const data = {};
-	for (const line of raw.split(/\r?\n/)) {
-		const trimmed = line.trim();
-		if (!trimmed || trimmed.startsWith("#")) continue;
-		const match = trimmed.match(/^([A-Za-z0-9_-]+):\s*(.*)$/);
-		if (!match) continue;
-		data[match[1]] = unquote(match[2]);
-	}
-	return data;
-}
-
-function unquote(value) {
-	const trimmed = value.trim();
-	if ((trimmed.startsWith('"') && trimmed.endsWith('"')) || (trimmed.startsWith("'") && trimmed.endsWith("'"))) {
-		return trimmed.slice(1, -1);
-	}
-	return trimmed;
-}
-
-function checkFrontmatter(frontmatter, errors) {
-	for (const field of ["title", "description", "lang"]) {
-		if (!frontmatter[field]) errors.push(`frontmatter: ${field} is required`);
-	}
-	if (frontmatter.heroImage && !isHttpUrl(frontmatter.heroImage) && !frontmatter.heroImage.startsWith("/")) {
-		errors.push("frontmatter: heroImage must be an absolute http(s) URL or root-relative path");
-	}
-}
-
-const ALLOWED_IMPORTS = new Set([
-	"@/components/mdx",
-	"@/components/cms",
-	"@/components/content",
-]);
-
-function checkImports(mdx, errors, warnings) {
-	const imports = [...mdx.matchAll(/^\s*import\s+[^'"]+['"]([^'"]+)['"];?\s*$/gm)].map((m) => m[1]);
-	for (const specifier of imports) {
-		if (![...ALLOWED_IMPORTS].some((allowed) => specifier.startsWith(allowed))) {
-			errors.push(`import: unsupported import "${specifier}"`);
-		}
-	}
-}
-
-const ALLOWED_COMPONENTS = new Set([
-	"ArticleHeader",
-	"Callout",
-	"CTA",
-	"FAQ",
-	"Image",
-	"ImageGrid",
-	"ProductCard",
-	"RelatedContent",
-	"Table",
-]);
-
-function checkComponents(mdx, errors) {
-	const tags = [...mdx.matchAll(/<([A-Z][A-Za-z0-9.]*)\b/g)].map((m) => m[1].split(".")[0]);
-	for (const tag of new Set(tags)) {
-		if (!ALLOWED_COMPONENTS.has(tag)) errors.push(`component: unsupported component <${tag}>`);
-	}
-}
-
-function checkLinks(mdx, warnings) {
-	const markdownLinks = [...mdx.matchAll(/\[[^\]]+\]\(([^)]+)\)/g)].map((m) => m[1]);
-	const hrefs = [...mdx.matchAll(/\bhref=["']([^"']+)["']/g)].map((m) => m[1]);
-	for (const url of [...markdownLinks, ...hrefs]) {
-		if (url.startsWith("#") || url.startsWith("/") || isHttpUrl(url) || url.startsWith("mailto:") || url.startsWith("tel:")) continue;
-		warnings.push(`link: review non-standard URL "${url}"`);
-	}
-}
-
-function checkImages(mdx, sourcePack, warnings) {
-	const urls = [
-		...[...mdx.matchAll(/!\[[^\]]*\]\(([^)]+)\)/g)].map((m) => m[1]),
-		...[...mdx.matchAll(/\bsrc=["']([^"']+)["']/g)].map((m) => m[1]),
-		...[...mdx.matchAll(/\bimage=["']([^"']+)["']/g)].map((m) => m[1]),
-		...[...mdx.matchAll(/\bheroImage:\s*["']?([^"'\n]+)["']?/g)].map((m) => m[1].trim()),
-	].filter((url) => isHttpUrl(url));
-	const allowed = collectAllowedImageUrls(sourcePack);
-	for (const url of new Set(urls)) {
-		if (allowed.size > 0 && !allowed.has(url)) {
-			warnings.push(`image: "${url}" is not present in source-pack image results`);
-		}
-	}
-}
-
-function collectAllowedImageUrls(sourcePack) {
-	const urls = new Set();
-	const add = (value) => {
-		if (typeof value === "string" && isHttpUrl(value)) urls.add(value);
-	};
-	const walk = (value) => {
-		if (Array.isArray(value)) return value.forEach(walk);
-		if (!value || typeof value !== "object") return;
-		for (const [key, child] of Object.entries(value)) {
-			if (/^(url|imageUrl|src|heroImage)$/i.test(key)) add(child);
-			walk(child);
-		}
-	};
-	walk(sourcePack?.images);
-	walk(sourcePack?.imageResults);
-	walk(sourcePack?.action?.target);
-	return urls;
-}
-
-async function checkMdxCompile(mdx, errors, warnings) {
-	try {
-		const mod = await import("@mdx-js/mdx");
-		await mod.compile(mdx, { jsx: true, development: false });
-	} catch (err) {
-		const message = err instanceof Error ? err.message : String(err);
-		if (message.includes("Cannot find package '@mdx-js/mdx'")) {
-			warnings.push("mdx: @mdx-js/mdx is not installed in this repo; skipped compiler parse check");
-			checkBalancedBraces(mdx, warnings);
-			return;
-		}
-		errors.push(`mdx: compile failed: ${message}`);
-	}
-}
-
-function checkBalancedBraces(mdx, warnings) {
-	const opens = (mdx.match(/{/g) ?? []).length;
-	const closes = (mdx.match(/}/g) ?? []).length;
-	if (opens !== closes) warnings.push(`mdx: brace count differs ({=${opens}, }=${closes}); review JSX expressions`);
-}
-
-function isHttpUrl(value) {
-	return /^https?:\/\//i.test(value);
-}
-
-function required(value, name) {
-	if (!value) throw new Error(`${name} is required`);
-	return value;
-}