agentsclimarketplace

Agent memory

Skill IgorGanapolsky/ThumbGate/skills/agent-memory

Give your agent persistent cross-session memory — recall what went wrong on similar tasks before starting, then capture whether the work succeeded or failed so repeated mistakes auto-promote into prevention rules. Runs locally via the ThumbGate MCP server (recall, capture_feedback, prevention_rules, feedback_stats) — no API key needed. Use when the user says "recall", "what went wrong last time", "remember this for next time", "capture feedback", "thumbs up/down on that", or at the start of a new task or session. Do NOT use when the user wants to force one specific mistake into an enforced block gate (use the thumbgate-guard skill), to list rules that already exist (thumbgate-rules), or to check ThumbGate wiring (thumbgate-doctor).From its SKILL.md

Install
npx -y skills add IgorGanapolsky/ThumbGate --skill agent-memory

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

3.8 KB, as published. Nobody here has run it

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

ToolWhat it does
recallSearch past feedback and prevention rules for current task
capture_feedbackRecord what worked or failed with structured context
prevention_rulesView auto-generated rules from repeated mistakes
feedback_statsApproval rate, trend analysis, top failure domains
feedback_summaryHuman-readable summary of recent signals

MCP Profiles

ProfileToolsUse case
essential5 core toolsDefault — start here
commerce6 tools + commerce_recallAgentic commerce agents
default12 toolsFull pipeline including DPO export

Set profile: THUMBGATE_MCP_PROFILE=essential npx thumbgate serve

How Prevention Rules Work

  1. Agent makes mistake A → you capture down feedback
  2. Agent makes mistake A again → you capture down feedback again
  3. System detects pattern → auto-generates prevention rule: "NEVER do A"
  4. Next session → recall returns the rule → agent follows it

This is the core value. The agent doesn't learn — but it reads the rules and follows them.

Links

Keep looking

Skills are one crate of 613,872. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.