docs: add README quickstart install links

.openclaw/index.js

@@ -1,23 +0,0 @@
-const SUPERPOWERS_GUIDANCE = `## Superpowers
-
-You have access to the Superpowers skill framework through OpenClaw's native skill system.
-Before responding to software development work, check whether one of those skills applies.
-If a Superpowers skill fits the task, invoke it before proceeding.
-
-Start with \`using-superpowers\` when you need the overall workflow. Common follow-on
-skills include \`brainstorming\`, \`writing-plans\`, \`test-driven-development\`,
-\`systematic-debugging\`, \`dispatching-parallel-agents\`, and
-\`verification-before-completion\`.
-
-When Superpowers instructions mention generic tools, use the closest native OpenClaw
-tool or workflow.`;
-
-export default {
-  id: "superpowers-openclaw",
-  name: "Superpowers for OpenClaw",
-  description: "Expose the Superpowers skill pack through OpenClaw's native plugin skill discovery.",
-
-  register(api) {
-    api.on("before_prompt_build", async () => ({ prependSystemContext: SUPERPOWERS_GUIDANCE }));
-  },
-};

.version-bump.json

@@ -4,7 +4,6 @@
     { "path": ".claude-plugin/plugin.json", "field": "version" },
     { "path": ".cursor-plugin/plugin.json", "field": "version" },
     { "path": ".codex-plugin/plugin.json", "field": "version" },
-    { "path": "openclaw.plugin.json", "field": "version" },
     { "path": ".claude-plugin/marketplace.json", "field": "plugins.0.version" },
     { "path": "gemini-extension.json", "field": "version" }
   ],

README.md

@@ -127,33 +127,6 @@ already use it in another harness.
 
 - Detailed docs: [docs/README.opencode.md](docs/README.opencode.md)
 
-### OpenClaw
-
-OpenClaw uses its native plugin system. Install Superpowers separately even if
-you already use it in another harness.
-
-- Clone and link the plugin:
-
-  ```bash
-  git clone https://github.com/obra/superpowers.git ~/.openclaw/vendor/superpowers
-  openclaw plugins install --link ~/.openclaw/vendor/superpowers --dangerously-force-unsafe-install
-  openclaw plugins enable superpowers-openclaw
-  openclaw gateway restart
-  ```
-
-  The override is required because OpenClaw scans the linked plugin directory
-  during native plugin install. It flags `skills/writing-skills/render-graphs.js`,
-  an existing Superpowers skill-authoring helper, because that script shells out
-  to Graphviz. The helper is not part of the OpenClaw runtime hook; review the
-  source before using the override.
-
-- Verify the plugin and skills are available:
-
-  ```bash
-  openclaw plugins info superpowers-openclaw --json
-  openclaw skills info using-superpowers --json
-  ```
-
 ### Cursor
 
 - In Cursor Agent chat, install from marketplace:

hooks/hooks-cursor.json

@@ -3,7 +3,7 @@
   "hooks": {
     "sessionStart": [
       {
-        "command": "./hooks/run-hook.cmd session-start"
+        "command": "./hooks/session-start"
       }
     ]
   }

openclaw.plugin.json

@@ -1,13 +0,0 @@
-{
-  "id": "superpowers-openclaw",
-  "name": "Superpowers for OpenClaw",
-  "version": "5.0.7",
-  "description": "Expose the Superpowers skill pack through OpenClaw's native plugin skill discovery.",
-  "skills": [
-    "./skills"
-  ],
-  "configSchema": {
-    "type": "object",
-    "additionalProperties": false
-  }
-}

package.json

@@ -2,10 +2,5 @@
   "name": "superpowers",
   "version": "5.0.7",
   "type": "module",
-  "main": ".opencode/plugins/superpowers.js",
-  "openclaw": {
-    "extensions": [
-      "./.openclaw/index.js"
-    ]
-  }
+  "main": ".opencode/plugins/superpowers.js"
 }

scripts/sync-to-codex-plugin.sh

@@ -52,7 +52,6 @@ EXCLUDES=(
   "/.gitattributes"
   "/.github/"
   "/.gitignore"
-  "/.openclaw/"
   "/.opencode/"
   "/.version-bump.json"
   "/.worktrees/"
@@ -65,7 +64,6 @@ EXCLUDES=(
   "/GEMINI.md"
   "/RELEASE-NOTES.md"
   "/gemini-extension.json"
-  "/openclaw.plugin.json"
   "/package.json"
 
   # Directories not shipped by canonical Codex plugins

skills/using-superpowers/references/codex-tools.md

@@ -6,7 +6,7 @@ Skills use Claude Code tool names. When you encounter these in a skill, use your
 |-----------------|------------------|
 | `Task` tool (dispatch subagent) | `spawn_agent` (see [Named agent dispatch](#named-agent-dispatch)) |
 | Multiple `Task` calls (parallel) | Multiple `spawn_agent` calls |
-| Task returns result | `wait_agent` |
+| Task returns result | `wait` |
 | Task completes automatically | `close_agent` to free slot |
 | `TodoWrite` (task tracking) | `update_plan` |
 | `Skill` tool (invoke a skill) | Skills load natively — just follow the instructions |
@@ -22,12 +22,7 @@ Add to your Codex config (`~/.codex/config.toml`):
 multi_agent = true
 ```
 
-This enables `spawn_agent`, `wait_agent`, and `close_agent` for skills like `dispatching-parallel-agents` and `subagent-driven-development`.
-
-Legacy note: Codex builds before `rust-v0.115.0` exposed spawned-agent
-waiting as `wait`. Current Codex uses `wait_agent` for spawned agents. The
-`wait` name now belongs to code-mode `exec/wait`, which resumes a yielded exec
-cell by `cell_id`; it is not the spawned-agent result tool.
+This enables `spawn_agent`, `wait`, and `close_agent` for skills like `dispatching-parallel-agents` and `subagent-driven-development`.
 
 ## Named agent dispatch
 

skills/using-superpowers/references/gemini-tools.md

@@ -14,29 +14,11 @@ Skills use Claude Code tool names. When you encounter these in a skill, use your
 | `Skill` tool (invoke a skill) | `activate_skill` |
 | `WebSearch` | `google_web_search` |
 | `WebFetch` | `web_fetch` |
-| `Task` tool (dispatch subagent) | `@agent-name` (see [Subagent support](#subagent-support)) |
+| `Task` tool (dispatch subagent) | No equivalent — Gemini CLI does not support subagents |
 
-## Subagent support
+## No subagent support
 
-Gemini CLI supports subagents natively via the `@` syntax. Use the built-in `@generalist` agent to dispatch any task — it has access to all tools and follows the prompt you provide.
-
-When a skill says to dispatch a named agent type, use `@generalist` with the full prompt from the skill's prompt template:
-
-| Skill instruction | Gemini CLI equivalent |
-|-------------------|----------------------|
-| `Task tool (superpowers:implementer)` | `@generalist` with the filled `implementer-prompt.md` template |
-| `Task tool (superpowers:spec-reviewer)` | `@generalist` with the filled `spec-reviewer-prompt.md` template |
-| `Task tool (superpowers:code-reviewer)` | `@code-reviewer` (bundled agent) or `@generalist` with the filled review prompt |
-| `Task tool (superpowers:code-quality-reviewer)` | `@generalist` with the filled `code-quality-reviewer-prompt.md` template |
-| `Task tool (general-purpose)` with inline prompt | `@generalist` with your inline prompt |
-
-### Prompt filling
-
-Skills provide prompt templates with placeholders like `{WHAT_WAS_IMPLEMENTED}` or `[FULL TEXT of task]`. Fill all placeholders and pass the complete prompt as the message to `@generalist`. The prompt template itself contains the agent's role, review criteria, and expected output format — `@generalist` will follow it.
-
-### Parallel dispatch
-
-Gemini CLI supports parallel subagent dispatch. When a skill asks you to dispatch multiple independent subagent tasks in parallel, request all of those `@generalist` or named subagent tasks together in the same prompt. Keep dependent tasks sequential, but do not serialize independent subagent tasks just to preserve a simpler history.
+Gemini CLI has no equivalent to Claude Code's `Task` tool. Skills that rely on subagent dispatch (`subagent-driven-development`, `dispatching-parallel-agents`) will fall back to single-session execution via `executing-plans`.
 
 ## Additional Gemini CLI tools
 

tests/claude-code/test-subagent-driven-development-integration.sh

@@ -135,7 +135,8 @@ EOF
 
 # Note: We use a longer timeout since this is integration testing
 # Use --allowed-tools to enable tool usage in headless mode
-PROMPT="Execute the implementation plan at docs/superpowers/plans/implementation-plan.md using the subagent-driven-development skill.
+# IMPORTANT: Run from superpowers directory so local dev skills are available
+PROMPT="Change to directory $TEST_PROJECT and then execute the implementation plan at docs/superpowers/plans/implementation-plan.md using the subagent-driven-development skill.
 
 IMPORTANT: Follow the skill exactly. I will be verifying that you:
 1. Read the plan once at the beginning
@@ -146,14 +147,9 @@ IMPORTANT: Follow the skill exactly. I will be verifying that you:
 
 Begin now. Execute the plan."
 
-PLUGIN_DIR=$(cd "$SCRIPT_DIR/../.." && pwd)
-
-# Run claude from inside the test project so its session JSONL lands in a
-# project-specific directory under ~/.claude/projects/, isolated from any
-# other concurrent claude sessions.
-echo "Running Claude (plugin-dir: $PLUGIN_DIR, cwd: $TEST_PROJECT)..."
+echo "Running Claude (output will be shown below and saved to $OUTPUT_FILE)..."
 echo "================================================================================"
-cd "$TEST_PROJECT" && timeout 1800 claude -p "$PROMPT" --plugin-dir "$PLUGIN_DIR" --allowed-tools=all --permission-mode bypassPermissions 2>&1 | tee "$OUTPUT_FILE" || {
+cd "$SCRIPT_DIR/../.." && timeout 1800 claude -p "$PROMPT" --allowed-tools=all --add-dir "$TEST_PROJECT" --permission-mode bypassPermissions 2>&1 | tee "$OUTPUT_FILE" || {
     echo ""
     echo "================================================================================"
     echo "EXECUTION FAILED (exit code: $?)"
@@ -165,17 +161,13 @@ echo ""
 echo "Execution complete. Analyzing results..."
 echo ""
 
-# Find the session transcript. Because we ran claude from $TEST_PROJECT (a
-# unique tmp dir), its sessions live in their own ~/.claude/projects/ folder
-# and we can pick the most-recent one without racing other concurrent sessions.
-# Resolve the real path because macOS mktemp returns /var/... but claude
-# normalizes it to /private/var/... when naming the project dir.
-TEST_PROJECT_REAL=$(cd "$TEST_PROJECT" && pwd -P)
-# Claude normalizes the cwd to a directory name by replacing every non-alphanumeric
-# character with `-` (so `_`, `.`, `/` all become `-`).
-SESSION_DIR="$HOME/.claude/projects/$(echo "$TEST_PROJECT_REAL" | sed 's|[^a-zA-Z0-9]|-|g')"
-# `|| true` prevents pipefail killing the script if ls gets SIGPIPE'd by head.
-SESSION_FILE=$(ls -t "$SESSION_DIR"/*.jsonl 2>/dev/null | head -1 || true)
+# Find the session transcript
+# Session files are in ~/.claude/projects/-<working-dir>/<session-id>.jsonl
+WORKING_DIR_ESCAPED=$(echo "$SCRIPT_DIR/../.." | sed 's/\//-/g' | sed 's/^-//')
+SESSION_DIR="$HOME/.claude/projects/$WORKING_DIR_ESCAPED"
+
+# Find the most recent session file (created during this test run)
+SESSION_FILE=$(find "$SESSION_DIR" -name "*.jsonl" -type f -mmin -60 2>/dev/null | sort -r | head -1)
 
 if [ -z "$SESSION_FILE" ]; then
     echo "ERROR: Could not find session transcript file"
@@ -202,9 +194,9 @@ else
 fi
 echo ""
 
-# Test 2: Subagents were used (Agent / Task tool — name varies by harness version)
+# Test 2: Subagents were used (Task tool)
 echo "Test 2: Subagents dispatched..."
-task_count=$(grep -cE '"name":"(Agent|Task)"' "$SESSION_FILE" || echo "0")
+task_count=$(grep -c '"name":"Task"' "$SESSION_FILE" || echo "0")
 if [ "$task_count" -ge 2 ]; then
     echo "  [PASS] $task_count subagents dispatched"
 else

tests/openclaw/test-native-plugin.sh

@@ -1,135 +0,0 @@
-#!/usr/bin/env bash
-# Test: OpenClaw native plugin package
-# Verifies the Superpowers repo exposes a native OpenClaw plugin contract.
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
-REPO_ROOT="$(cd "$SCRIPT_DIR/../.." && pwd)"
-
-cd "$REPO_ROOT"
-
-echo "=== Test: OpenClaw native plugin package ==="
-
-echo "Test 1: Checking OpenClaw manifest..."
-node <<'NODE'
-import fs from "node:fs";
-
-const manifestPath = "openclaw.plugin.json";
-if (!fs.existsSync(manifestPath)) {
-  throw new Error(`${manifestPath} is missing`);
-}
-
-const manifest = JSON.parse(fs.readFileSync(manifestPath, "utf8"));
-const pkg = JSON.parse(fs.readFileSync("package.json", "utf8"));
-if (manifest.id !== "superpowers-openclaw") {
-  throw new Error(`unexpected manifest id: ${manifest.id}`);
-}
-if (manifest.version !== pkg.version) {
-  throw new Error(`manifest version ${manifest.version} must match package version ${pkg.version}`);
-}
-if (!Array.isArray(manifest.skills) || manifest.skills.length !== 1 || manifest.skills[0] !== "./skills") {
-  throw new Error(`unexpected skills declaration: ${JSON.stringify(manifest.skills)}`);
-}
-if (
-  !manifest.configSchema ||
-  manifest.configSchema.type !== "object" ||
-  manifest.configSchema.additionalProperties !== false ||
-  manifest.configSchema.properties !== undefined
-) {
-  throw new Error(`manifest must declare an empty object configSchema: ${JSON.stringify(manifest.configSchema)}`);
-}
-if ("entrypoint" in manifest) {
-  throw new Error("OpenClaw entrypoints belong in package.json openclaw.extensions, not openclaw.plugin.json");
-}
-if ("hooks" in manifest) {
-  throw new Error("OpenClaw hook registration belongs in the runtime entrypoint, not openclaw.plugin.json");
-}
-NODE
-echo "  [PASS] Manifest declares plugin skills correctly"
-
-echo "Test 2: Checking package OpenClaw extension metadata..."
-node <<'NODE'
-import fs from "node:fs";
-
-const pkg = JSON.parse(fs.readFileSync("package.json", "utf8"));
-if (pkg.name !== "superpowers") {
-  throw new Error(`package name changed unexpectedly: ${pkg.name}`);
-}
-if (pkg.type !== "module") {
-  throw new Error("OpenClaw runtime should preserve the repo's ESM package mode");
-}
-const extensions = pkg.openclaw?.extensions;
-if (!Array.isArray(extensions) || extensions.length !== 1 || extensions[0] !== "./.openclaw/index.js") {
-  throw new Error(`unexpected openclaw.extensions: ${JSON.stringify(extensions)}`);
-}
-NODE
-echo "  [PASS] package.json points OpenClaw at the runtime entrypoint"
-
-echo "Test 3: Checking runtime entrypoint syntax and hook behavior..."
-node --check .openclaw/index.js
-node <<'NODE'
-const module = await import(new URL("./.openclaw/index.js", import.meta.url));
-const plugin = module.default;
-
-if (!plugin || plugin.id !== "superpowers-openclaw" || typeof plugin.register !== "function") {
-  throw new Error("OpenClaw runtime must default-export a plugin with register(api)");
-}
-
-const hooks = [];
-const api = {
-  pluginConfig: {},
-  rootDir: process.cwd(),
-  logger: { warn(message) { throw new Error(`unexpected warning: ${message}`); } },
-  on(name, handler) {
-    hooks.push({ name, handler });
-  },
-};
-
-plugin.register(api);
-if (hooks.length !== 1 || hooks[0].name !== "before_prompt_build") {
-  throw new Error(`unexpected hooks: ${JSON.stringify(hooks.map((hook) => hook.name))}`);
-}
-
-const result = await hooks[0].handler();
-if (!result?.prependSystemContext?.includes("Superpowers")) {
-  throw new Error("before_prompt_build hook must inject Superpowers guidance");
-}
-if (result.prependSystemContext.includes("~/.openclaw/skills")) {
-  throw new Error("native plugin guidance must not advertise managed skill symlinks");
-}
-NODE
-echo "  [PASS] Runtime entrypoint registers the prompt hook"
-
-echo "Test 4: Checking README install flow..."
-if ! grep -q 'openclaw plugins install --link ~/.openclaw/vendor/superpowers' README.md; then
-  echo "  [FAIL] README must document linked OpenClaw plugin install"
-  exit 1
-fi
-if ! grep -q -- '--dangerously-force-unsafe-install' README.md; then
-  echo "  [FAIL] README must document OpenClaw scanner override required by existing helper scripts"
-  exit 1
-fi
-if ! grep -q 'skills/writing-skills/render-graphs.js' README.md; then
-  echo "  [FAIL] README must explain the exact Superpowers helper that triggers OpenClaw's scanner"
-  exit 1
-fi
-if ! grep -q 'not part of the OpenClaw runtime hook' README.md; then
-  echo "  [FAIL] README must explain that the flagged helper is not part of the OpenClaw runtime hook"
-  exit 1
-fi
-if ! grep -q 'openclaw plugins enable superpowers-openclaw' README.md; then
-  echo "  [FAIL] README must document enabling the OpenClaw plugin"
-  exit 1
-fi
-if ! grep -q 'openclaw gateway restart' README.md; then
-  echo "  [FAIL] README must document restarting the gateway"
-  exit 1
-fi
-if grep -q 'OPENCLAW_SKILLS_DIR\|ln -s\|AGENTS-snippet\|.openclaw/INSTALL.md' README.md; then
-  echo "  [FAIL] README must not use the legacy symlink/snippet wrapper or extra install doc"
-  exit 1
-fi
-echo "  [PASS] README documents native plugin commands"
-
-echo ""
-echo "=== OpenClaw native plugin tests passed ==="