Add shadow-skill entrypoint

.env.example

@@ -1,6 +1,6 @@
-# Privacy Router provider configuration.
+# shadow skill provider configuration.
 #
-# The vendor-neutral privacy-router skill enables providers independently.
+# shadow skill enables privacy providers independently.
 # Missing credentials disable only that provider; `list-providers` will still
 # show the disabled reason.
 #

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2026 HoudiniSkill contributors
+Copyright (c) 2026 shadow skill contributors
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

package.json

@@ -1,7 +1,7 @@
 {
-  "name": "houdiniskill",
+  "name": "shadow-skill",
   "version": "0.1.0",
-  "description": "Privacy-preserving cross-chain crypto swap Agent Skill, powered by HoudiniSwap Partner API v2. Compatible with Claude Code, OpenClaw, Hermes (Nous Research), Pi (earendil-works), Cursor and any tool supporting the SKILL.md standard.",
+  "description": "Comprehensive privacy-preserving cross-chain crypto transfer and swap Agent Skill with provider-neutral routing. HoudiniSwap is the first bundled provider adapter. Compatible with Claude Code, OpenClaw, Hermes (Nous Research), Pi, Cursor and any tool supporting the SKILL.md standard.",
   "type": "module",
   "private": false,
   "license": "MIT",
@@ -18,6 +18,8 @@
     "pi-coding-agent",
     "cursor",
     "skill-md",
+    "shadow-skill",
+    "privacy-router",
     "houdiniswap",
     "crypto",
     "defi",
@@ -29,14 +31,14 @@
     "test": "node scripts/run-tests.mjs",
     "test:coverage": "node scripts/run-tests.mjs --coverage",
     "smoke": "node --test --test-reporter=spec skills/houdiniswap/tests/_client.test.mjs skills/houdiniswap/tests/list-chains.test.mjs skills/privacy-router/tests/registry.test.mjs skills/privacy-router/tests/router.test.mjs",
-    "lint:syntax": "node --check skills/houdiniswap/scripts/_client.mjs && node --check skills/privacy-router/lib/router.mjs && node --check scripts/router/list-providers.mjs"
+    "lint:syntax": "node --check skills/houdiniswap/scripts/_client.mjs && node --check skills/privacy-router/lib/router.mjs && node --check scripts/router/list-providers.mjs && node --check skills/shadow-skill/scripts/list-providers.mjs"
   },
   "repository": {
     "type": "git",
-    "url": "git+https://github.com/REPLACE_ME/HoudiniSkill.git"
+    "url": "git+https://github.com/REPLACE_ME/shadow-skill.git"
   },
   "bugs": {
-    "url": "https://github.com/REPLACE_ME/HoudiniSkill/issues"
+    "url": "https://github.com/REPLACE_ME/shadow-skill/issues"
   },
-  "homepage": "https://github.com/REPLACE_ME/HoudiniSkill#readme"
+  "homepage": "https://github.com/REPLACE_ME/shadow-skill#readme"
 }

scripts/install.ps1

@@ -1,12 +1,12 @@
 <#
 .SYNOPSIS
-  HoudiniSkill installer (Windows PowerShell).
+  shadow skill installer (Windows PowerShell).
 
 .DESCRIPTION
   - Verifies Node.js >= 22 is installed
   - Copies .env.example to .env (preserving existing .env)
   - Runs the offline smoke test suite
-  - Symlinks/junctions skills/houdiniswap and skills/privacy-router into the standard agent skill paths
+  - Symlinks/junctions skills/shadow-skill, skills/privacy-router, and skills/houdiniswap into the standard agent skill paths
     (Claude Code / OpenClaw / Hermes / Pi / Cursor) when they exist on this machine
   - Verifies the deployment is readable end-to-end
 
@@ -44,7 +44,7 @@ $RequiredMajor = 22
 $RepoRoot = Resolve-Path (Join-Path $PSScriptRoot '..')
 Set-Location $RepoRoot
 
-$SkillNames = @('houdiniswap', 'privacy-router')
+$SkillNames = @('shadow-skill', 'privacy-router', 'houdiniswap')
 
 $AgentTargets = @(
     @{ Label = 'Claude Code (global)'; Parent = (Join-Path $env:USERPROFILE '.claude\skills') },
@@ -63,7 +63,7 @@ function Write-Info($msg) { Write-Host "  $msg" }
 function Write-Warn($msg) { Write-Host "  $msg" -ForegroundColor Yellow }
 function Write-Fail($msg) { Write-Host "  $msg" -ForegroundColor Red }
 
-Write-Bold "==> HoudiniSkill installer"
+Write-Bold "==> shadow skill installer"
 Write-Info "Repo root:   $RepoRoot"
 Write-Info "Skills:      $($SkillNames -join ', ')"
 ""
@@ -128,7 +128,7 @@ if (-not $LinkOnly) {
     } else {
         Copy-Item .env.example .env
         Write-Info "Created .env from .env.example."
-        Write-Info "Edit .env to add HOUDINISWAP_API_KEY + HOUDINISWAP_API_SECRET (optional)."
+        Write-Info "Edit .env to add HOUDINISWAP_API_KEY + HOUDINISWAP_API_SECRET for the bundled HoudiniSwap provider (optional)."
     }
     ""
 
@@ -207,8 +207,9 @@ Write-Bold "==> Project-scoped install (optional)"
 
     cd <your-project>
     New-Item -ItemType Directory -Force .claude\skills | Out-Null
-    New-Item -ItemType Junction -Path ".claude\skills\houdiniswap" -Target "$(Join-Path $RepoRoot 'skills\houdiniswap')"
+    New-Item -ItemType Junction -Path ".claude\skills\shadow-skill" -Target "$(Join-Path $RepoRoot 'skills\shadow-skill')"
     New-Item -ItemType Junction -Path ".claude\skills\privacy-router" -Target "$(Join-Path $RepoRoot 'skills\privacy-router')"
+    New-Item -ItemType Junction -Path ".claude\skills\houdiniswap" -Target "$(Join-Path $RepoRoot 'skills\houdiniswap')"
 
 "@ | Write-Host
 
@@ -234,18 +235,18 @@ if ($linkFailures -gt 0) {
 
   What this did:
 
-    - Linked skills/houdiniswap/ and skills/privacy-router/ into every agent's global skills directory
+    - Linked skills/shadow-skill/, skills/privacy-router/, and skills/houdiniswap/ into every agent's global skills directory
       that already exists on this machine (Claude Code / OpenClaw / Hermes /
       Pi / Cursor).
-    - The skill auto-loads .env from the cloned repo root (via _client.mjs),
-      so changes to .env take effect for any agent that picks up the skill.
+    - The bundled HoudiniSwap provider auto-loads .env from the cloned repo root (via _client.mjs),
+      so changes to .env take effect for any agent that picks up shadow skill.
 
   Next steps:
 
     - Restart your agent (Claude Desktop / Code CLI / OpenClaw / Hermes /
       Pi / Cursor) so it re-scans its skills directory.
     - Try a prompt:
-        "Use the privacy-router skill to find a private cross-chain route for 26 USDT (eth) -> USDT (bsc)."
+        "Use shadow-skill to find a private cross-chain route for 26 USDT (eth) -> USDT (bsc)."
 
   Verify connectivity to the real API (optional, no funds spent):
 

scripts/install.sh

@@ -1,10 +1,10 @@
 #!/usr/bin/env bash
-# HoudiniSkill installer (POSIX)
+# shadow skill installer (POSIX)
 #
 # - Verifies Node.js >= 22 is installed
 # - Copies .env.example to .env (preserving existing .env)
 # - Runs the offline smoke test suite
-# - Symlinks skills/houdiniswap and skills/privacy-router into the standard agent skill paths so
+# - Symlinks skills/shadow-skill, skills/privacy-router, and skills/houdiniswap into the standard agent skill paths so
 #   Claude Code / OpenClaw / Hermes / Pi / Cursor / Codex / etc. discover it
 # - Verifies the deployment is readable end-to-end
 #
@@ -56,7 +56,7 @@ fail() { printf '\033[31m  %s\033[0m\n' "$*" >&2; }
 
 # ---------- Source-of-truth paths ------------------------------------------
 
-SKILL_NAMES=("houdiniswap" "privacy-router")
+SKILL_NAMES=("shadow-skill" "privacy-router" "houdiniswap")
 
 # Agent skill destinations. Format: "<label>|<dir>"
 # Each entry's parent must already exist (we never auto-create $HOME/.foo
@@ -69,7 +69,7 @@ AGENT_TARGETS=(
   "Cursor (global)|$HOME/.cursor/skills"
 )
 
-bold "==> HoudiniSkill installer"
+bold "==> shadow skill installer"
 info "Repo root:   $REPO_ROOT"
 info "Skills:      ${SKILL_NAMES[*]}"
 echo
@@ -134,7 +134,7 @@ EOF
   else
     cp .env.example .env
     info "Created .env from .env.example."
-    info "Edit .env to add HOUDINISWAP_API_KEY + HOUDINISWAP_API_SECRET (optional)."
+    info "Edit .env to add HOUDINISWAP_API_KEY + HOUDINISWAP_API_SECRET for the bundled HoudiniSwap provider (optional)."
   fi
   echo
 
@@ -214,12 +214,14 @@ cat <<EOF
 
     cd <your-project>
     mkdir -p .claude/skills
-    ln -snf "$REPO_ROOT/skills/houdiniswap" .claude/skills/houdiniswap
+    ln -snf "$REPO_ROOT/skills/shadow-skill" .claude/skills/shadow-skill
     ln -snf "$REPO_ROOT/skills/privacy-router" .claude/skills/privacy-router
+    ln -snf "$REPO_ROOT/skills/houdiniswap" .claude/skills/houdiniswap
     # or:
     mkdir -p .cursor/skills
-    ln -snf "$REPO_ROOT/skills/houdiniswap" .cursor/skills/houdiniswap
+    ln -snf "$REPO_ROOT/skills/shadow-skill" .cursor/skills/shadow-skill
     ln -snf "$REPO_ROOT/skills/privacy-router" .cursor/skills/privacy-router
+    ln -snf "$REPO_ROOT/skills/houdiniswap" .cursor/skills/houdiniswap
 
 EOF
 
@@ -246,18 +248,18 @@ cat <<'EOF'
 
   What this did:
 
-    - Linked skills/houdiniswap/ and skills/privacy-router/ into every agent's global skills directory
+    - Linked skills/shadow-skill/, skills/privacy-router/, and skills/houdiniswap/ into every agent's global skills directory
       that already exists on this machine (Claude Code / OpenClaw / Hermes /
       Pi / Cursor).
-    - The skill auto-loads .env from the cloned repo root (via _client.mjs),
-      so changes to .env take effect for any agent that picks up the skill.
+    - The bundled HoudiniSwap provider auto-loads .env from the cloned repo root (via _client.mjs),
+      so changes to .env take effect for any agent that picks up shadow skill.
 
   Next steps:
 
     - Restart your agent (Claude Desktop / Code CLI / OpenClaw / Hermes /
       Pi / Cursor) so it re-scans its skills directory.
     - Try a prompt:
-        "Use the privacy-router skill to find a private cross-chain route for 26 USDT (eth) -> USDT (bsc)."
+        "Use shadow-skill to find a private cross-chain route for 26 USDT (eth) -> USDT (bsc)."
 
   Verify connectivity to the real API (optional, no funds spent):
 

scripts/verify-deployment.sh

@@ -1,12 +1,13 @@
 #!/usr/bin/env bash
-# Verify that this skill is correctly deployed for an agent.
+# Verify that shadow skill is correctly deployed for an agent.
 #
 # Checks performed (in order):
-#   1. The skill's source dir contains SKILL.md with a valid frontmatter
+#   1. Each skill source dir contains SKILL.md with a valid frontmatter
 #      (name + description).
 #   2. At least one agent skills directory contains a deployed copy/link of
-#      the skills at <agent-dir>/houdiniswap/SKILL.md and
-#      <agent-dir>/privacy-router/SKILL.md.
+#      the skills at <agent-dir>/shadow-skill/SKILL.md,
+#      <agent-dir>/privacy-router/SKILL.md, and
+#      <agent-dir>/houdiniswap/SKILL.md.
 #   3. From the deployed location, `node` can locate the source code's .env
 #      via _client.mjs's loader (or via process.env). Reports the resolved
 #      auth mode without printing the secrets.
@@ -24,7 +25,7 @@
 set -euo pipefail
 
 REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
-SKILL_NAMES=("houdiniswap" "privacy-router")
+SKILL_NAMES=("shadow-skill" "privacy-router" "houdiniswap")
 
 QUIET=0
 EXTRA_TARGETS=()
@@ -107,7 +108,7 @@ DEFAULT_TARGETS=(
 # When --target is given we only check those; otherwise the defaults.
 # Explicit targets may be either an agent skills parent directory
 # (~/.claude/skills) or one concrete skill directory
-# (~/.claude/skills/houdiniswap).
+# (~/.claude/skills/shadow-skill).
 TARGET_LABELS=()
 TARGET_CHECKS=()
 add_target_checks() {
@@ -160,6 +161,21 @@ for check in "${TARGET_CHECKS[@]}"; do
   info "OK  $dest_skill_md"
 
   if [ "$skill_name" != "houdiniswap" ]; then
+    entrypoint_mjs="$dest/scripts/list-providers.mjs"
+    if [ ! -f "$entrypoint_mjs" ]; then
+      warn "$dest is missing scripts/list-providers.mjs (broken deployment?)"
+      verify_failures=$((verify_failures + 1))
+      continue
+    fi
+    entrypoint_out="$(
+      cd "$dest" && HOUDINISWAP_SKIP_DOTENV=1 node scripts/list-providers.mjs 2>&1
+    )" || {
+      fail "Entrypoint probe failed inside $dest"
+      fail "  $entrypoint_out"
+      verify_failures=$((verify_failures + 1))
+      continue
+    }
+    info "    entrypoint: OK"
     continue
   fi
 

skills/houdiniswap/SKILL.md

@@ -6,9 +6,10 @@ license: MIT
 
 # HoudiniSwap Agent Skill
 
-This skill turns natural-language swap intents into HoudiniSwap Partner API v2
-calls. It is suited for privacy-conscious cross-chain swaps where the user
-wants to disconnect the source wallet from the destination wallet on-chain.
+This is the direct low-level HoudiniSwap provider skill for shadow skill.
+Prefer `shadow-skill` when the user wants the agent to choose among providers.
+Use this skill when the user explicitly needs HoudiniSwap-specific commands or
+provider behavior.
 
 ## Prerequisites
 
@@ -301,7 +302,7 @@ Pure-local introspection of the chain RPC pool. Reports each registered
 chain's adapter, env var name, built-in pool size, whether an override is
 currently active, and the effective ordered URL list `verify-tx-onchain`
 would try. No API auth needed; never makes a network call. Use it to answer
-"does HoudiniSkill have a built-in RPC for chain X?" or "did my custom
+"does shadow skill have a built-in RPC for chain X?" or "did my custom
 override actually get picked up?". Exit `1` if the requested `--chain` isn't
 registered with any adapter (the JSON output still includes a `note` hint).
 

skills/houdiniswap/references/architecture.md

@@ -1,8 +1,9 @@
 # Architecture
 
-HoudiniSkill is an installable Agent Skill package for HoudiniSwap, not a
-long-running service. The runtime surface is a set of small Node.js CLI scripts
-that an agent can call after reading `SKILL.md`.
+shadow skill is an installable Agent Skill package for privacy-preserving
+transfers and cross-chain swaps, not a long-running service. HoudiniSwap is the
+first bundled provider adapter. The runtime surface is a set of small Node.js
+CLI scripts that an agent can call after reading `SKILL.md`.
 
 ## System Overview
 
@@ -35,6 +36,7 @@ validation, JSON output, and exit codes.
 flowchart LR
   subgraph Docs["Docs and agent entrypoint"]
     R[README.md]
+    SS[skills/shadow-skill/SKILL.md]
     S[skills/houdiniswap/SKILL.md]
     Ref[references/*.md]
   end
@@ -131,7 +133,7 @@ The adapter layer implements the chain-specific mechanics:
 sequenceDiagram
   participant User
   participant Agent
-  participant CLI as HoudiniSkill CLI
+  participant CLI as shadow skill CLI
   participant API as HoudiniSwap API
   participant Chain as Chain RPC/REST
 

skills/houdiniswap/references/status-codes.md

@@ -51,7 +51,7 @@ This applies to every script in `skills/houdiniswap/scripts/`.
 | `2`  | Usage / argument error             | Missing required flag, bad value, malformed input  |
 | `3`  | Network error                      | Timeout, connection refused, abort                 |
 | `4`  | Auth not configured                | `AuthMissingError` from `_client.mjs`              |
-| `70` | Internal error (unexpected)        | Anything not derived from `HoudiniSkillError`      |
+| `70` | Internal error (unexpected)        | Anything not derived from `ShadowSkillError`       |
 
 ## How `wait-for-order` decides when to stop