opencode-skills-antigravity 1.0.9 → 1.0.10

package/bundled-skills/docs/users/agent-overload-recovery.md

@@ -0,0 +1,54 @@
+# Antigravity Recovery for Context Overload and Truncation
+
+Use this guide when Antigravity loads too many skills for the current task and starts failing with truncation, context, or trajectory-conversion errors.
+
+Typical symptoms:
+
+- the agent crashes only when a large skills folder is present
+- the error mentions truncation, context conversion, or a trajectory/message that cannot be converted
+- the problem shows up more often on large repositories or long-running tasks
+
+## Linux and macOS fast path
+
+Use the activation scripts to keep the full library archived while exposing only the bundles or skills you need in the live Antigravity directory.
+
+1. Fully close Antigravity.
+2. Clone this repository somewhere local if you do not already have a clone.
+3. Run the activation script from the cloned repository.
+
+Examples:
+
+```bash
+./scripts/activate-skills.sh "Web Wizard" "Integration & APIs"
+./scripts/activate-skills.sh --clear
+./scripts/activate-skills.sh brainstorming systematic-debugging
+```
+
+What the script does:
+
+- syncs the repository `skills/` tree into `~/.gemini/antigravity/skills_library`
+- preserves your full library in the backing store
+- activates only the requested bundles or skill ids into `~/.gemini/antigravity/skills`
+- `--clear` archives the current live directory first, then restores the selected set
+
+Optional environment overrides:
+
+```bash
+AG_BASE_DIR=/custom/antigravity ./scripts/activate-skills.sh --clear Essentials
+AG_REPO_SKILLS_DIR=/path/to/repo/skills ./scripts/activate-skills.sh brainstorming
+```
+
+## Windows recovery
+
+If Antigravity is stuck in a restart loop on Windows, use the Windows-specific recovery guide instead:
+
+- [windows-truncation-recovery.md](windows-truncation-recovery.md)
+
+That guide covers the browser/app storage cleanup needed when the host keeps reopening the same broken session.
+
+## Prevention tips
+
+- start with 3-5 skills from a bundle instead of exposing the full library at once
+- use bundle activation before opening very large repositories
+- keep role-specific stacks active and archive the rest
+- if a host stores broken session state, clear that host state before restoring a smaller active set

package/bundled-skills/docs/users/faq.md

@@ -154,6 +154,14 @@ It includes:
 - the default Antigravity Windows paths to back up first
 - an optional batch script adapted from [issue #274](https://github.com/sickn33/antigravity-awesome-skills/issues/274)
 
+### I hit context overload on Linux or macOS. What should I do?
+
+If Antigravity becomes unstable only when the full skills library is active, switch to the activation flow instead of exposing every skill at once:
+
+- [agent-overload-recovery.md](agent-overload-recovery.md)
+
+That guide shows how to run `scripts/activate-skills.sh` from a cloned copy of this repository so only the bundles or skill ids you need stay active in `~/.gemini/antigravity/skills`.
+
 ### How do I update skills?
 
 Navigate to your skills directory and pull the latest changes:

package/bundled-skills/docs/users/getting-started.md

@@ -1,4 +1,4 @@
-# Getting Started with Antigravity Awesome Skills (V8.5.0)
+# Getting Started with Antigravity Awesome Skills (V8.6.0)
 
 **New here? This guide will help you supercharge your AI Agent in 5 minutes.**
 
@@ -142,6 +142,9 @@ A: Yes! Use the **@skill-creator** skill to build your own.
 **Q: What if Antigravity on Windows gets stuck in a truncation crash loop?**
 A: Follow the recovery steps in [windows-truncation-recovery.md](windows-truncation-recovery.md). It explains which Antigravity storage folders to back up and clear, and includes an optional batch helper adapted from [issue #274](https://github.com/sickn33/antigravity-awesome-skills/issues/274).
 
+**Q: What if Antigravity overloads on Linux or macOS when too many skills are active?**
+A: Use the activation flow in [agent-overload-recovery.md](agent-overload-recovery.md). It shows how to run `scripts/activate-skills.sh` from a cloned repo so you can keep the full library archived and activate only the bundles or skills you need in the live Antigravity directory.
+
 **Q: Is this free?**
 A: Yes. Original code and tooling are MIT-licensed, and original documentation/non-code written content is CC BY 4.0. See [../../LICENSE](../../LICENSE) and [../../LICENSE-CONTENT](../../LICENSE-CONTENT).
 

package/bundled-skills/docs/users/usage.md

@@ -40,7 +40,7 @@ Bundles are **recommended lists** of skills grouped by role. They help you decid
 
 ❌ Separate installations  
 ❌ Different download commands  
-❌ Something you need to "activate"
+❌ Something most users need to activate during normal install
 
 ### Example: The "Web Wizard" Bundle
 

package/bundled-skills/gdb-cli/SKILL.md

@@ -0,0 +1,239 @@
+---
+name: gdb-cli
+description: "GDB debugging assistant for AI agents - analyze core dumps, debug live processes, investigate crashes and deadlocks with source code correlation"
+category: development
+risk: critical
+source: community
+date_added: "2026-03-22"
+author: Cerdore
+tags:
+- debugging
+- gdb
+- core-dump
+- crash-analysis
+- c++
+- c
+tools:
+- claude-code
+- cursor
+- gemini-cli
+- codex-cli
+- antigravity
+---
+
+# GDB Debugging Assistant
+
+## Overview
+
+A GDB debugging skill designed for AI agents. Combines **source code analysis** with **runtime state inspection** using gdb-cli to provide intelligent debugging assistance for C/C++ programs.
+
+## When to Use This Skill
+
+- Analyze core dumps or crash dumps
+- Debug running processes with GDB attach
+- Investigate crashes, deadlocks, or memory issues
+- Get intelligent debugging assistance with source code context
+- Debug multi-threaded applications
+
+## Do Not Use This Skill When
+
+- The task is unrelated to C/C++ debugging
+- The user needs general-purpose assistance without debugging
+- No GDB is available (GDB 9.0+ with Python support required)
+
+## Prerequisites
+
+```bash
+# Install gdb-cli
+pip install gdb-cli
+
+# Or from GitHub
+pip install git+https://github.com/Cerdore/gdb-cli.git
+
+# Verify GDB has Python support
+gdb -nx -q -batch -ex "python print('OK')"
+```
+
+**Requirements:**
+- Python 3.6.8+
+- GDB 9.0+ with Python support enabled
+- Linux OS
+
+## How It Works
+
+### Step 1: Initialize Debug Session
+
+**For core dump analysis:**
+```bash
+gdb-cli load --binary <binary_path> --core <core_path> [--gdb-path <gdb_path>]
+```
+
+**For live process debugging:**
+```bash
+gdb-cli attach --pid <pid> [--binary <binary_path>]
+```
+
+**Output:** A session_id like `"session_id": "a1b2c3"`. Store this for subsequent commands.
+
+### Step 2: Gather Initial Information
+
+```bash
+SESSION="<session_id>"
+
+# List all threads
+gdb-cli threads -s $SESSION
+
+# Get backtrace (with local variables)
+gdb-cli bt -s $SESSION --full
+
+# Get registers
+gdb-cli registers -s $SESSION
+```
+
+### Step 3: Correlate Source Code (CRITICAL)
+
+For each frame in the backtrace:
+1. **Extract frame info**: `{file}:{line} in {function}`
+2. **Read source context**: Get ±20 lines around the crash point
+3. **Get local variables**: `gdb-cli locals-cmd -s $SESSION --frame <N>`
+4. **Analyze**: Correlate code logic with variable values
+
+**Example correlation:**
+```
+Frame #0: process_data() at src/worker.c:87
+Source code shows:
+  85: Node* node = get_node(id);
+  86: if (node == NULL) return;
+  87: node->data = value;  <- Crash here
+
+Variables show:
+  node = 0x0 (NULL)
+
+Analysis: The NULL check on line 86 didn't catch the issue.
+```
+
+### Step 4: Deep Investigation
+
+```bash
+# Examine variables
+gdb-cli eval-cmd -s $SESSION "variable_name"
+gdb-cli eval-cmd -s $SESSION "ptr->field"
+gdb-cli ptype -s $SESSION "struct_name"
+
+# Memory inspection
+gdb-cli memory -s $SESSION "0x7fffffffe000" --size 64
+
+# Disassembly
+gdb-cli disasm -s $SESSION --count 20
+
+# Check all threads (for deadlock analysis)
+gdb-cli thread-apply -s $SESSION bt --all
+
+# View shared libraries
+gdb-cli sharedlibs -s $SESSION
+```
+
+### Step 5: Session Management
+
+```bash
+# List active sessions
+gdb-cli sessions
+
+# Check session status
+gdb-cli status -s $SESSION
+
+# Stop session (cleanup)
+gdb-cli stop -s $SESSION
+```
+
+## Common Debugging Patterns
+
+### Pattern: Null Pointer Dereference
+
+**Indicators:**
+- Crash on memory access instruction
+- Pointer variable is 0x0
+
+**Investigation:**
+```bash
+gdb-cli registers -s $SESSION  # Check RIP
+gdb-cli eval-cmd -s $SESSION "ptr"  # Check pointer value
+```
+
+### Pattern: Deadlock
+
+**Indicators:**
+- Multiple threads stuck in lock functions
+- `pthread_mutex_lock` in backtrace
+
+**Investigation:**
+```bash
+gdb-cli thread-apply -s $SESSION bt --all
+# Look for circular wait patterns
+```
+
+### Pattern: Memory Corruption
+
+**Indicators:**
+- Crash in malloc/free
+- Garbage values in variables
+
+**Investigation:**
+```bash
+gdb-cli memory -s $SESSION "&variable" --size 128
+gdb-cli registers -s $SESSION
+```
+
+## Examples
+
+### Example 1: Core Dump Analysis
+
+```bash
+# Load core dump
+gdb-cli load --binary ./myapp --core /tmp/core.1234
+
+# Get crash location
+gdb-cli bt -s a1b2c3 --full
+
+# Examine crash frame
+gdb-cli locals-cmd -s a1b2c3 --frame 0
+```
+
+### Example 2: Live Process Debugging
+
+```bash
+# Attach to stuck server
+gdb-cli attach --pid 12345
+
+# Check all threads
+gdb-cli threads -s b2c3d4
+
+# Get all backtraces
+gdb-cli thread-apply -s b2c3d4 bt --all
+```
+
+## Best Practices
+
+- Always read source code before drawing conclusions from variable values
+- Use `--range` for pagination on large thread counts or deep backtraces
+- Use `ptype` to understand complex data structures before examining values
+- Check all threads for multi-threaded issues
+- Cross-reference types with source code definitions
+
+## Security & Safety Notes
+
+- This skill requires GDB access to processes and core dumps
+- Attaching to processes may require appropriate permissions (sudo, ptrace_scope)
+- Core dumps may contain sensitive data - handle with care
+- Only debug processes you have authorization to analyze
+
+## Related Skills
+
+- `@systematic-debugging` - General debugging methodology
+- `@test-driven-development` - Write tests before implementation
+
+## Links
+
+- **Repository**: https://github.com/Cerdore/gdb-cli
+- **PyPI**: https://pypi.org/project/gdb-cli/
+- **Documentation**: https://github.com/Cerdore/gdb-cli#readme

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-skills-antigravity",
-  "version": "1.0.9",
+  "version": "1.0.10",
   "description": "OpenCode CLI plugin that automatically downloads and keeps Antigravity Awesome Skills up to date.",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",

package/bundled-skills/goldrush-api/SKILL.md

@@ -1,109 +0,0 @@
----
-name: goldrush-api
-description: "Query blockchain data across 100+ chains: wallet balances, token prices, transactions, DEX pairs, and real-time OHLCV streams via the GoldRush API by Covalent."
-category: blockchain
-risk: safe
-source: community
-date_added: "2026-03-17"
-author: covalenthq
-tags: [blockchain, crypto, web3, api, defi, wallet, multi-chain]
-tools: [claude, cursor, gemini, codex, copilot]
----
-
-# GoldRush API
-
-## Overview
-GoldRush by Covalent provides blockchain data across 100+ chains through a unified REST API, real-time WebSocket streams, a CLI, and an x402 pay-per-request proxy. This skill enables AI agents to query wallet balances, token prices, transaction history, NFT holdings, and DEX pair data without building chain-specific integrations.
-
-## When to Use This Skill
-- Retrieving wallet token balances or total portfolio value across any chain
-- Fetching transaction history or decoded event logs for an address
-- Getting current or historical token prices (USD or native)
-- Monitoring DEX pairs, liquidity events, and real-time OHLCV candles via WebSocket
-- Building block explorers, portfolio dashboards, tax tools, or DeFi analytics
-- Accessing on-chain data with no signup via x402 pay-per-request
-
-## How It Works
-
-### Step 1: Get credentials
-Sign up at https://goldrush.dev for a free API key. For agent-native no-signup access, use the x402 proxy instead.
-
-### Step 2: Set your API key
-```bash
-export GOLDRUSH_API_KEY="your_api_key_here"
-```
-
-### Step 3: Query data
-
-**REST API (most endpoints):**
-```bash
-# Wallet token balances on Ethereum
-curl -H "Authorization: Bearer $GOLDRUSH_API_KEY"   "https://api.covalenthq.com/v1/eth-mainnet/address/0xADDRESS/balances_v2/"
-```
-
-**CLI (quick terminal queries):**
-```bash
-npx @covalenthq/goldrush-cli balances --chain eth-mainnet --address 0xADDRESS
-```
-
-**SDK (in code):**
-```javascript
-import GoldRushClient from "@covalenthq/client-sdk";
-const client = new GoldRushClient(process.env.GOLDRUSH_API_KEY);
-const resp = await client.BalanceService.getTokenBalancesForWalletAddress(
-  "eth-mainnet", "0xADDRESS"
-);
-```
-
-## Examples
-
-### Example 1: Token Balances
-```bash
-curl -H "Authorization: Bearer $GOLDRUSH_API_KEY"   "https://api.covalenthq.com/v1/eth-mainnet/address/0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045/balances_v2/"
-```
-
-### Example 2: Token Price History
-```bash
-curl -H "Authorization: Bearer $GOLDRUSH_API_KEY"   "https://api.covalenthq.com/v1/pricing/historical_by_addresses_v2/eth-mainnet/USD/0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48/"
-```
-
-### Example 3: Transaction History
-```bash
-curl -H "Authorization: Bearer $GOLDRUSH_API_KEY"   "https://api.covalenthq.com/v1/eth-mainnet/address/0xADDRESS/transactions_v3/"
-```
-
-### Example 4: Real-time OHLCV via WebSocket
-```javascript
-// Stream live price candles for a token pair
-const ws = new WebSocket("wss://streaming.covalenthq.com/v1/eth-mainnet/ohlcv");
-ws.on("message", (data) => console.log(JSON.parse(data)));
-```
-
-## Best Practices
-✅ Use chain slugs: `eth-mainnet`, `matic-mainnet`, `base-mainnet`, `bsc-mainnet` — full list at https://goldrush.dev/docs/networks
-✅ Store API key in `GOLDRUSH_API_KEY` env var — never hardcode
-✅ Use WebSocket streams for real-time data rather than polling REST
-✅ Use SDK cursor pagination for large result sets
-❌ Don't use x402 for high-volume use cases — get a standard API key instead
-❌ Don't use chain IDs (e.g., `1`) — use chain slugs (e.g., `eth-mainnet`)
-
-## Security & Safety Notes
-- API key in `GOLDRUSH_API_KEY` environment variable only
-- x402 payments use USDC on Base — set spending limits before autonomous agent use
-- Read-only data API — no write operations, no transaction signing
-
-## Common Pitfalls
-
-**Problem:** 401 Unauthorized  
-**Solution:** Ensure API key is in `Authorization: Bearer` header, not query string
-
-**Problem:** `chain_name not found`  
-**Solution:** Use chain slug format — see https://goldrush.dev/docs/networks
-
-**Problem:** Empty results for new wallet  
-**Solution:** Some endpoints require on-chain activity; new wallets with no transactions return empty arrays, not errors
-
-## Related Skills
-- @goldrush-streaming-api — real-time WebSocket DEX pair and OHLCV streams
-- @goldrush-x402 — pay-per-request blockchain data without API key
-- @goldrush-cli — terminal-first blockchain data queries