Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add IgorGanapolsky/ThumbGate --skill agent-memorygit clone --depth 1 https://github.com/IgorGanapolsky/ThumbGateWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/igorganapolsky/thumbgate/agent-memory)<a href="https://agentmods.dev/skills/igorganapolsky/thumbgate/agent-memory"><img src="https://agentmods.dev/badge/skills/igorganapolsky/thumbgate/agent-memory/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/igorganapolsky/thumbgate/agent-memory"><img src="https://agentmods.dev/badge/skills/igorganapolsky/thumbgate/agent-memory.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 15 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 82 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00162 | $0.00873 |
| Opus 5.5 | $0.00065 | $0.00349 |
| Sonnet 5.5 | $0.00032 | $0.00175 |
| Haiku 4.5 | $0.00016 | $0.00087 |
Grade A, and why
Agent Memory scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Memory
Give your agent persistent memory across sessions. Before starting any task, recall what went wrong last time. After completing work, capture whether it succeeded or failed. Prevention rules are auto-generated from repeated mistakes.
Setup
Add the MCP server (one-time):
claude mcp add thumbgate -- npx -y thumbgate serve
No API key needed. All data stays local.
When to Use
- Starting a new task or session — recall past context first
- After completing work that succeeded or failed — capture feedback
- When the agent keeps making the same mistake — check prevention rules
Workflow
Step 1: Recall past context (do this FIRST on every task)
Call the recall MCP tool with a description of your current task. The tool returns:
- Past feedback relevant to this task (vector similarity search)
- Active prevention rules (auto-generated from repeated failures)
- Recent feedback summary with approval rate
Read the prevention rules carefully. These are patterns that failed before — follow them.
Step 2: Do your work
Complete the task as normal. Keep track of what you did and whether it worked.
Step 3: Capture feedback
Call the capture_feedback MCP tool:
If succeeded:
- signal:
up - context: What worked and why
- tags: Category labels
If failed:
- signal:
down - context: What you were trying to do
- whatWentWrong: Specific failure description
- whatToChange: How to avoid this next time
- tags: Category labels
Vague feedback like "it failed" will be rejected. Be specific.
Step 4: Check improvement (optional)
Call the feedback_stats MCP tool to see approval rate, top failure domains, and whether the agent is trending better or worse.
Available MCP Tools
| Tool | What it does |
|---|---|
recall |
Search past feedback and prevention rules for current task |
capture_feedback |
Record what worked or failed with structured context |
prevention_rules |
View auto-generated rules from repeated mistakes |
feedback_stats |
Approval rate, trend analysis, top failure domains |
feedback_summary |
Human-readable summary of recent signals |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 98 lines · 162 tokens per session scan A e7edbb837f0a
Agent Memory is a skill published in the GitHub repository IgorGanapolsky/ThumbGate (29 stars, last pushed today), licensed MIT. It adds 162 tokens to every session and 873 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
llamaguard
Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.
cordis-plugin-sofagent-evolve
A sofagent add-on for recording lessons from completed work and refining future agent instructions.
llamaguard
Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.
persist
Write stable learnings, decisions, and patterns to durable storage like CLAUDE.md or knowledge files. Use when saving project decisions, recording patterns, or updating long-term memory.
recall
Retrieve prior decisions, rationale, and learned patterns from memory to apply consistently. Use when needing context from previous interactions, looking up past decisions, or ensuring consistency with prior reasoning.
llamaguard
Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.