anymorph 0.2.6 → 0.5.0

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "anymorph",
-	"version": "0.2.6",
+	"version": "0.5.0",
 	"description": "Check your brand's AI visibility across ChatGPT, Perplexity, Gemini, and more",
 	"type": "module",
 	"private": false,
@@ -29,6 +29,7 @@
 		"ora": "^8.0.0"
 	},
 	"devDependencies": {
+		"@anymorph/shared-contracts": "file:../shared-contracts",
 		"@types/node": "^22.0.0",
 		"esbuild": "^0.25.0",
 		"tsx": "^4.0.0",

package/dist/skillpacks/geo/skills/brand-owned-diagnosis/SKILL.md

@@ -1,54 +0,0 @@
----
-name: brand-owned-diagnosis
-description: Use when diagnosing customer-owned discovered pages for GEO strategy actions, owned-site visibility gaps, content coverage issues, or technical page fixes.
----
-
-# Brand-Owned Diagnosis
-
-Use this skill for customer-owned pages that are not Anymorph-generated GEO pages.
-
-The output is a compact diagnosis plus action candidates for `assetType="brand_owned"`.
-
-## Scope
-
-Diagnose:
-
-- Existing brand-owned page coverage.
-- Content clarity, proof, comparison, and entity gaps.
-- Technical or crawl blockers on customer-owned pages.
-- GSC, SERP, AI response, and competitor gaps tied to a known owned URL.
-
-Do not propose:
-
-- Generated GEO page work.
-- Third-party outreach or external profile work.
-- New page generation when an existing generated GEO page is the right surface.
-
-## Workflow
-
-Read these references before producing proposals:
-
-- `references/workflow.md` for the full brand-owned product workflow, candidate rules, root causes, action families, scoring, and no-action cases.
-- `references/diagnosis-contract.md` for the exact runtime diagnosis contract and artifact submission shape.
-
-1. Read shared run context and the brand-owned candidate slice.
-2. Use `list_geo_strategy_candidates` before broad SQL.
-3. Use `get_geo_intent_diagnostic` for intent-level explanation.
-4. Read page content or source only when the recommendation depends on the body or technical implementation.
-5. Return accepted proposals, rejected candidates, and no-action rationale.
-
-## Output Contract
-
-Return proposals that can be merged into final GEO actions.
-
-Each proposal should include:
-
-- intent or proposed intent.
-- existing page URL or page id when known.
-- root cause.
-- operation family.
-- priority and confidence.
-- evidence references.
-- concise change brief.
-
-Use no action when the owned page is not the bottleneck.

package/dist/skillpacks/geo/skills/brand-owned-diagnosis/agents/openai.yaml

@@ -1,4 +0,0 @@
-interface:
-  display_name: "Brand-Owned Diagnosis"
-  short_description: "Diagnose customer-owned pages for GEO strategy actions."
-  default_prompt: "Use $brand-owned-diagnosis to diagnose owned-page GEO gaps for this workspace."

package/dist/skillpacks/geo/skills/brand-owned-diagnosis/references/diagnosis-contract.md

@@ -1,96 +0,0 @@
-# GEO Strategy: Brand-Owned Diagnosis
-
-You diagnose customer-owned discovered pages.
-
-## Scope
-
-Work only on pages where the tenant/customer site owns the page and the page is not an Anymorph-generated GEO page.
-
-Cases:
-
-- page exists but visibility is low
-- page is cited but the brand is not mentioned
-- brand is mentioned but not recommended
-
-Do not propose GEO page generation unless the orchestrator explicitly asks you to compare update-vs-create.
-
-## Evidence Order
-
-Use the cheapest sufficient evidence.
-
-1. Common run context and candidate rows from the orchestrator.
-2. `list_geo_strategy_candidates` for the first brand-owned candidate slice.
-3. `get_geo_intent_diagnostic` for one intent before deeper checks.
-4. `get_page_insights`, `get_page_visibility`, or `list_ai_responses` for page/prompt evidence.
-5. `get_page_content` or `get_page_source_code` only when content or technical diagnosis depends on the page body.
-6. `web_scrape` only for a specific competitor URL that changes the recommendation.
-
-Do not scrape broadly.
-
-## Root Cause Buckets
-
-Allowed buckets:
-
-- `intent_coverage_gap`
-- `content_depth_gap`
-- `entity_clarity_gap`
-- `generic_content_gap`
-- `proof_gap`
-- `comparison_gap`
-- `technical_issue`
-- `authority_gap`
-- `prompt_mismatch`
-- `not_a_problem`
-- `insufficient_evidence`
-
-Use `authority_gap` only with external citation evidence or repeated competitor recommendation reasons.
-
-## Action Families
-
-Allowed families:
-
-- `refresh_existing_page`
-- `retarget_existing_page`
-- `improve_entity_clarity`
-- `add_proof`
-- `add_comparison_coverage`
-- `fix_technical_issue`
-- `no_action`
-
-## Output
-
-Produce one `brand_owned` diagnosis object in this shape.
-
-Do not write intermediate diagnosis JSON files directly.
-
-Do not include `workspaceId`, `runId`, `channel`, final action id, or final `assetType`; the orchestrator can derive them.
-
-Use this shape:
-
-```json
-{
-  "summary": "Short diagnosis.",
-  "proposals": [
-    {
-      "intentId": "intent_123",
-      "candidateType": "page_exists_low_visibility",
-      "target": { "pageId": "page_123", "url": "https://example.com/page" },
-      "rootCauses": [
-        { "bucket": "proof_gap", "confidence": 0.72, "evidence": ["Specific evidence summary"] }
-      ],
-      "actionFamily": "add_proof",
-      "operations": ["Add customer proof near the answer block"],
-      "priority": "high",
-      "confidence": "medium",
-      "evidence": [
-        { "type": "signal", "ref": "geo_recommendation_quadrants:intent_123", "summary": "Mentioned but not recommended dominates." }
-      ],
-      "whyNow": "Why this matters this week.",
-      "risks": ["Uncertainty or dependency"]
-    }
-  ],
-  "rejected": [
-    { "ref": "intent_456", "reason": "Suppressed by recent action mask." }
-  ]
-}
-```

package/dist/skillpacks/geo/skills/brand-owned-diagnosis/references/workflow.md

@@ -1,191 +0,0 @@
-# Brand-Owned Diagnosis Workflow
-
-## Role
-
-The `brand_owned` channel diagnosis diagnoses and proposes actions for customer-owned discovered pages.
-
-It works on pages where:
-
-- `source = discovered`
-- the page belongs to the tenant/customer site
-- the action would improve an existing natural page or recommend a customer-site content action
-
-## Non-Goals
-
-- Do not iterate generated GEO pages.
-- Do not decide final routes for generated pages.
-- Do not propose third-party outreach.
-- Do not perform deep URL tree planning in MVP.
-- Do not require `PageContentLabel`.
-
-## Inputs
-
-- candidate rows routed by the orchestrator
-- tenant `agent/BRAND.md`
-- tenant `agent/STRATEGY.md`
-- tenant `agent/LEARNINGS.md`
-- relevant rows from:
-  - `geo_pages`
-  - `geo_intent_coverage`
-  - `geo_recommendation_quadrants`
-  - `geo_page_intent_visibility_gaps`
-  - `geo_page_performance`
-  - `geo_strategy_actions_recent`
-- page scrape when needed
-- AI answer samples when needed
-- competitor page scrape when needed
-
-## Workflow
-
-### 1. Filter Candidates
-
-Keep candidates where:
-
-- page source is `discovered`
-- page has an intent
-- intent is weak, or quadrant state indicates a fixable issue
-
-Drop candidates when:
-
-- page is not indexable and the tenant strategy forbids technical page edits
-- recent action mask suppresses the same action
-- evidence is too thin to justify an action
-
-### 2. Classify Candidate Case
-
-Assign one primary case.
-
-#### A. Page Exists But Visibility Is Low
-
-Use when:
-
-- discovered page exists
-- `current_visibility` is low
-- quadrant is mostly `absent`, `cited_unnamed`, or `mentioned_not_recommended`
-
-Diagnosis questions:
-
-1. Does the page directly answer the intent?
-2. Is coverage deep enough for representative prompts?
-3. Is the writing weaker than the most-cited or most-recommended competitor page?
-4. Is there a visible technical issue?
-5. Is authority the likely bottleneck, or only a weak hypothesis?
-
-#### B. Cited But Not Mentioned
-
-Use when:
-
-- our page/source is cited
-- brand is not clearly mentioned
-
-Diagnosis questions:
-
-1. Is the page clearly tied to the tenant brand?
-2. Do title, H1, intro, byline, publisher, schema, and about/proof signals make the entity clear?
-3. Is the content too generic to make the brand worth mentioning?
-
-#### C. Mentioned But Not Recommended
-
-Use when:
-
-- tenant brand is mentioned
-- competitors are recommended more often
-
-Diagnosis questions:
-
-1. Which brands were recommended?
-2. What reasons did the AI answer give or imply?
-3. What content-level weakness exists on the tenant page?
-
-### 3. Gather Evidence
-
-Use the cheapest sufficient evidence.
-
-Default evidence:
-
-- view rows
-- page title/H1/path/status
-- quadrant rates
-- page performance summary
-- AI answer excerpt
-
-Fetch page scrape when:
-
-- diagnosis depends on content coverage, proof, comparison, or entity clarity
-
-Fetch competitor page scrape when:
-
-- the answer repeatedly recommends or cites a competitor page
-- the proposed action depends on competitor comparison
-
-Do not scrape everything.
-
-### 4. Pick Root Cause Buckets
-
-Allowed buckets:
-
-- `intent_coverage_gap`
-- `content_depth_gap`
-- `entity_clarity_gap`
-- `generic_content_gap`
-- `proof_gap`
-- `comparison_gap`
-- `technical_issue`
-- `authority_gap`
-- `prompt_mismatch`
-- `not_a_problem`
-- `insufficient_evidence`
-
-Use `authority_gap` only when supported by external citation evidence or repeated competitor recommendation reasons.
-
-### 5. Route Action
-
-Allowed action families:
-
-- `refresh_existing_page`
-- `retarget_existing_page`
-- `improve_entity_clarity`
-- `add_proof`
-- `add_comparison_coverage`
-- `fix_technical_issue`
-- `no_action`
-
-Prefer update actions over create actions when an existing discovered page clearly owns the intent.
-
-### 6. Output
-
-Return a compact diagnosis for each accepted action:
-
-- candidate type
-- page URL/pageId
-- intentId
-- root cause buckets with confidence
-- recommended action family
-- concrete operations
-- evidence references
-- why other actions were rejected
-
-## Examples
-
-### Covered But Weak
-
-If a discovered page exists, visibility is low, and the page only gives generic education while competitor pages provide comparison criteria and proof:
-
-- root cause: `comparison_gap`, `proof_gap`
-- action family: `add_comparison_coverage`
-- operation: add comparison criteria and customer proof to the existing page
-
-### Cited But Not Mentioned
-
-If the page is cited but the AI answer does not name the tenant brand:
-
-- root cause: `entity_clarity_gap`
-- action family: `improve_entity_clarity`
-- operation: strengthen brand naming, publisher schema, and product/company relationship in the answer block
-
-## Guardrails
-
-- Do not propose creating a generated GEO page from this channel diagnosis unless the orchestrator explicitly routes no-page candidates here.
-- Do not over-index on page type labels.
-- Do not call content "thin" without page scrape or word-count/section evidence.
-- Do not diagnose authority from low visibility alone.

package/dist/skillpacks/geo/skills/geo-generating-actions/SKILL.md

@@ -1,100 +0,0 @@
----
-name: geo-generating-actions
-description: Use when generating recurring or on-demand GEO strategy actions from tenant memory, run context, channel diagnoses, and evidence-backed opportunity candidates.
----
-
-# GEO Generating Actions
-
-Use this skill to orchestrate GEO action generation.
-
-This skill owns the final decision. Channel diagnosis should be handled by product diagnosis skills, then merged and judged here.
-
-## Channel Skills
-
-Use these skills for channel-specific diagnosis:
-
-- `$brand-owned-diagnosis` for customer-owned pages.
-- `$geo-pages-diagnosis` for generated GEO pages.
-- `$third-party-diagnosis` for Earn and external authority work.
-
-## Required Reference
-
-Read `references/orchestrator.workflow.md` before final action selection. It contains the full production workflow: context construction, channel invocation order, proposal merge rules, conflict resolution, no-action discipline, and final artifact contract.
-
-## Workflow
-
-1. Read host-created run files and stable tenant memory.
-2. Build one compact shared run context.
-3. Invoke or apply the three channel diagnosis skills.
-4. Read channel outputs and reject weak, duplicate, or conflicting proposals.
-5. Prefer no action over low-evidence action.
-6. In local CLI mode, write `agent/runs/{runId}/actions.json` directly.
-7. Write rationale and status files only under the current run path.
-8. Update durable tenant memory only when the run changes stable strategy or repeated learnings.
-
-## Local CLI Mode
-
-When running inside a tenant repo, this skill expects a prepared run directory.
-Do not invent a `runId` or create fake input files.
-
-Use the CLI to prepare the same filesystem contract that remote runners use.
-
-```bash
-anymorph --version
-anymorph status
-anymorph geo init --repo .
-anymorph geo doctor --repo .
-anymorph geo prepare <workspace-domain-or-id> --repo .
-```
-
-`geo init` installs the bundled skills and local memory scaffold. It does not
-create a strategy run. `geo prepare` fetches the workspace run package from the
-backend and creates the run files.
-
-`geo prepare` creates:
-
-- `agent/runs/{runId}/manifest.json`
-- `agent/runs/{runId}/context.json`
-- `agent/runs/{runId}/intents.json`
-- `agent/runs/{runId}/crawl_logs.json`
-- `agent/runs/{runId}/status.json`
-
-`intents.json` comes from `signals.intents` in the backend prepare response.
-It is the same source used by the remote GEO strategy runner.
-
-Inspect prepared intents when needed:
-
-```bash
-anymorph geo intents {runId} --repo .
-```
-
-Then run this skill against that `runId`. Use MCP tools only to enrich the
-prepared context or inspect deeper evidence. Do not use MCP calls to replace
-the run package.
-
-For local CLI runs, there is no remote runtime artifact submission tool. The
-final artifact contract is the filesystem contract:
-
-- write `agent/runs/{runId}/actions.json`
-- write `agent/runs/{runId}/rationale.md`
-- update `agent/runs/{runId}/status.json`
-
-After writing artifacts, validate them with:
-
-```bash
-anymorph geo validate {runId} --repo .
-```
-
-Only use runtime artifact tools when the host environment explicitly provides
-them. In ordinary local tenant repo work, the filesystem artifacts above are the
-source of truth.
-
-## Final Decision Rules
-
-- Emit at most 20 actions.
-- Use only `create` or `update` operations.
-- Use `brand_owned` only for customer-owned discovered pages.
-- Use `geo_page` only for generated or managed GEO page work.
-- Use `third_party` only for off-page Earn work.
-- Every action must include either an existing `intentId` or a non-null `proposedIntent`.
-- Third-party actions must name the exact domain and, when known, the exact publish surface.

package/dist/skillpacks/geo/skills/geo-generating-actions/agents/openai.yaml

@@ -1,4 +0,0 @@
-interface:
-  display_name: "GEO Generating Actions"
-  short_description: "Generate final GEO strategy actions from channel diagnoses."
-  default_prompt: "Use $geo-generating-actions to generate final GEO strategy actions for this workspace."