feat: add Tool Use Guardian + RecallMax (free agent skills) (#298)
* feat: add Tool Use Guardian + RecallMax — free agent reliability & memory skills Two free skills from the Genesis Agent Marketplace: - **Tool Use Guardian**: Intelligent tool-call reliability wrapper. Monitors, retries, fixes, and learns from tool failures. Auto-recovers from truncated JSON, timeouts, rate limits, and mid-chain failures. - **RecallMax**: God-tier long-context memory. Injects 500K-1M clean tokens, auto-summarizes with tone/intent preservation, compresses 14-turn history into 800 high-density tokens. Both are free forever and work with Claude, Codex, Gemini, Copilot, Cursor, Windsurf, Antigravity, and any SKILL.md-compatible agent. Install: npx skills add christopherlhammer11-ai/tool-use-guardian npx skills add christopherlhammer11-ai/recallmax * docs: add source credits for genesis skills --------- Co-authored-by: Mia Hammer <miahammer@Mac.attlocal.net> Co-authored-by: sck_0 <samujackson1337@gmail.com>
This commit is contained in:
christopherlhammer11-ai
committed by
GitHub
GitHub
parent
c0a9b7a363
commit
16ff84afa1
@@ -431,6 +431,8 @@ This collection would not be possible without the incredible work of the Claude
|
||||
- **[Xquik-dev/x-twitter-scraper](https://github.com/Xquik-dev/x-twitter-scraper)**: X (Twitter) data platform — tweet search, user lookup, follower extraction, engagement metrics, giveaway draws, monitoring, webhooks, 19 extraction tools, MCP server.
|
||||
- **[shmlkv/dna-claude-analysis](https://github.com/shmlkv/dna-claude-analysis)**: Personal genome analysis toolkit — Python scripts analyzing raw DNA data across 17 categories (health risks, ancestry, pharmacogenomics, nutrition, psychology, etc.) with terminal-style single-page HTML visualization.
|
||||
- **[AlmogBaku/debug-skill](https://github.com/AlmogBaku/debug-skill)**: Interactive debugger skill for AI agents — breakpoints, stepping, variable inspection, and stack traces via the `dap` CLI. Supports Python, Go, Node.js/TypeScript, Rust, and C/C++.
|
||||
- **[christopherlhammer11-ai/tool-use-guardian](https://github.com/christopherlhammer11-ai/tool-use-guardian)**: Source for the Tool Use Guardian skill — tool-call reliability wrapper with retries, recovery, and failure classification.
|
||||
- **[christopherlhammer11-ai/recallmax](https://github.com/christopherlhammer11-ai/recallmax)**: Source for the RecallMax skill — long-context memory, summarization, and conversation compression for agents.
|
||||
|
||||
### Inspirations
|
||||
|
||||
|
||||
77
skills/recallmax/SKILL.md
Normal file
77
skills/recallmax/SKILL.md
Normal file
@@ -0,0 +1,77 @@
|
||||
---
|
||||
name: recallmax
|
||||
description: "FREE — God-tier long-context memory for AI agents. Injects 500K-1M clean tokens, auto-summarizes with tone/intent preservation, compresses 14-turn history into 800 tokens."
|
||||
category: memory
|
||||
risk: safe
|
||||
source: community
|
||||
date_added: "2026-03-13"
|
||||
author: christopherlhammer11-ai
|
||||
tags: [memory, context, rag, summarization, compression, long-context, agent-infrastructure]
|
||||
tools: [claude, cursor, codex, gemini, copilot, windsurf, antigravity, grok]
|
||||
---
|
||||
|
||||
# RecallMax — God-Tier Long-Context Memory
|
||||
|
||||
## Overview
|
||||
|
||||
RecallMax enhances AI agent memory capabilities dramatically. Inject 500K to 1M clean tokens of external context without hallucination drift. Auto-summarize conversations while preserving tone, sarcasm, and intent. Compress multi-turn histories into high-density token sequences.
|
||||
|
||||
Free forever. Built by the Genesis Agent Marketplace.
|
||||
|
||||
## Install
|
||||
|
||||
```bash
|
||||
npx skills add christopherlhammer11-ai/recallmax
|
||||
```
|
||||
|
||||
## When to Use This Skill
|
||||
|
||||
- Use when your agent loses context in long conversations (50+ turns)
|
||||
- Use when injecting large RAG/external documents into agent context
|
||||
- Use when you need to compress conversation history without losing meaning
|
||||
- Use when fact-checking claims across a long thread
|
||||
- Use for any agent that needs to remember everything
|
||||
|
||||
## How It Works
|
||||
|
||||
### Step 1: Context Injection
|
||||
|
||||
RecallMax cleanly injects external context (documents, RAG results, prior conversations) into the agent's working memory. Unlike naive concatenation, it:
|
||||
- Deduplicates overlapping content
|
||||
- Preserves source attribution
|
||||
- Prevents hallucination drift from context pollution
|
||||
|
||||
### Step 2: Adaptive Summarization
|
||||
|
||||
As conversations grow, RecallMax automatically summarizes older turns while preserving:
|
||||
- **Tone** — sarcasm, formality, urgency
|
||||
- **Intent** — what the user actually wants vs. what they said
|
||||
- **Key facts** — numbers, names, decisions, commitments
|
||||
- **Emotional register** — frustration, excitement, confusion
|
||||
|
||||
### Step 3: History Compression
|
||||
|
||||
Compress a 14-turn conversation history into ~800 high-density tokens that retain full semantic meaning. The compressed output can be re-expanded if needed.
|
||||
|
||||
### Step 4: Fact Verification
|
||||
|
||||
Built-in cross-reference checks for controversial or ambiguous claims within the conversation context. Flags contradictions and unsupported assertions.
|
||||
|
||||
## Best Practices
|
||||
|
||||
- ✅ Use RecallMax at the start of long-running agent sessions
|
||||
- ✅ Enable auto-summarization for conversations beyond 20 turns
|
||||
- ✅ Use compression before hitting context window limits
|
||||
- ✅ Let the fact verifier run on high-stakes outputs
|
||||
- ❌ Don't inject unvetted external content without dedup
|
||||
- ❌ Don't skip summarization and rely on raw truncation
|
||||
|
||||
## Related Skills
|
||||
|
||||
- `@tool-use-guardian` - Tool-call reliability wrapper (also free from Genesis Marketplace)
|
||||
|
||||
## Links
|
||||
|
||||
- **Repo:** https://github.com/christopherlhammer11-ai/recallmax
|
||||
- **Marketplace:** https://genesis-node-api.vercel.app
|
||||
- **Browse skills:** https://genesis-marketplace.vercel.app
|
||||
84
skills/tool-use-guardian/SKILL.md
Normal file
84
skills/tool-use-guardian/SKILL.md
Normal file
@@ -0,0 +1,84 @@
|
||||
---
|
||||
name: tool-use-guardian
|
||||
description: "FREE — Intelligent tool-call reliability wrapper. Monitors, retries, fixes, and learns from tool failures. Auto-recovers from truncated JSON, timeouts, rate limits, and mid-chain failures."
|
||||
category: reliability
|
||||
risk: safe
|
||||
source: community
|
||||
date_added: "2026-03-13"
|
||||
author: christopherlhammer11-ai
|
||||
tags: [reliability, tool-use, error-handling, retries, recovery, agent-infrastructure]
|
||||
tools: [claude, cursor, codex, gemini, copilot, windsurf, antigravity]
|
||||
---
|
||||
|
||||
# Tool Use Guardian
|
||||
|
||||
## Overview
|
||||
|
||||
The reliability wrapper every AI agent needs. Monitors tool calls, auto-retries failures, fixes truncated responses, and learns which tools are unreliable — so you never lose your chain of thought.
|
||||
|
||||
Free forever. Built by the Genesis Agent Marketplace.
|
||||
|
||||
## Install
|
||||
|
||||
```bash
|
||||
npx skills add christopherlhammer11-ai/tool-use-guardian
|
||||
```
|
||||
|
||||
## When to Use This Skill
|
||||
|
||||
- Use when tool calls return truncated or malformed JSON
|
||||
- Use when APIs timeout or rate-limit your agent mid-task
|
||||
- Use when a multi-step chain breaks partway through
|
||||
- Use when you need automatic retry logic without writing it yourself
|
||||
- Use for any agent workflow that depends on external tool reliability
|
||||
|
||||
## How It Works
|
||||
|
||||
### Step 1: Pre-Call Validation
|
||||
|
||||
Before every tool call, Guardian validates:
|
||||
- Required parameters are present and correctly typed
|
||||
- The tool is not marked as "unreliable" from previous failures
|
||||
- Request size is within known limits
|
||||
|
||||
### Step 2: Failure Classification
|
||||
|
||||
When a tool call fails, Guardian classifies the failure into one of 9 categories:
|
||||
|
||||
| Failure Type | Recovery Action |
|
||||
|---|---|
|
||||
| Truncated JSON | Re-fetch with pagination or smaller chunks |
|
||||
| API Timeout | Retry once with simpler request, then decompose |
|
||||
| Rate Limit (429) | Exponential backoff, max 3 retries |
|
||||
| Auth Expired | Flag for user intervention |
|
||||
| Mid-chain Break | Resume from last successful checkpoint |
|
||||
| Error-as-200 | Detect `{"error": "..."}` disguised as success |
|
||||
| Schema Mismatch | Attempt auto-coercion, warn if lossy |
|
||||
| Network Failure | Retry with jitter, max 2 attempts |
|
||||
| Unknown Error | Log full context, escalate to user |
|
||||
|
||||
### Step 3: Chain Protection
|
||||
|
||||
For multi-step tool chains, Guardian maintains checkpoints. If step 4 of 7 fails, it resumes from step 4 — never restarts from scratch.
|
||||
|
||||
### Step 4: Learning
|
||||
|
||||
Guardian tracks failure patterns per tool. After 3+ failures of the same type, it marks the tool as unreliable and suggests alternatives.
|
||||
|
||||
## Best Practices
|
||||
|
||||
- ✅ Let Guardian wrap all external tool calls automatically
|
||||
- ✅ Review Guardian's reliability reports to identify flaky tools
|
||||
- ✅ Use checkpoint recovery for long chains
|
||||
- ❌ Don't disable retry logic for rate-limited APIs
|
||||
- ❌ Don't ignore repeated failure warnings
|
||||
|
||||
## Related Skills
|
||||
|
||||
- `@recallmax` - Long-context memory enhancement (also free from Genesis Marketplace)
|
||||
|
||||
## Links
|
||||
|
||||
- **Repo:** https://github.com/christopherlhammer11-ai/tool-use-guardian
|
||||
- **Marketplace:** https://genesis-node-api.vercel.app
|
||||
- **Browse skills:** https://genesis-marketplace.vercel.app
|
||||
Reference in New Issue
Block a user