Rlhf Feedback Loop

ThumbGate 👍 👎

ThumbGate is the self-improving pre-action firewall for AI coding agents

AI coding agents repeat mistakes — and one wrong tool call can wipe a directory, leak a key, or push broken code.

---

What it does

ThumbGate is the local-first Pre-Action Checks engine for AI coding agents. It runs in the PreToolUse hook to evaluate the proposed tool call before execution — so costly mistakes can be caught before they happen.

ThumbGate GitHub star growth is measured with GitHub's privacy-safe GET /repos/{owner}/{repo}/stargazers/history endpoint (weekly counts, no stargazer identities). Run npm run stars:history -- --fixture tests/fixtures/github-star-history.json --json for the local proof. Stars are not npm installs and not revenue. The live GitHub Marketplace Action is ThumbGate Agent Governance (uses: IgorGanapolsky/ThumbGate@v1).

Usage over star count. Evaluate ThumbGate from the install path and live usage badges above (npx thumbgate init, Marketplace uses:, npm weekly downloads, GitHub clones), not from whether the repo has twenty stars or twenty thousand. No pitch deck is required. We do not farm GitHub profile badges (no YOLO-merge of protected main, no 5-minute Issue close theater, no fake Co-authored-by). Galaxy Brain needs real accepted answers in Discussions Q&A. npm run github:achievements -- --fixture tests/fixtures/github-achievements.json --json inventories what is already earned vs what we refuse to farm.

Who it's for

ThumbGate is for operators whose AI coding agents can leak a secret or destroy a checkout before a human sees the tool call (Claude Code, Cursor, Codex, Gemini CLI, MCP). Discovery should reach those operators — not a star campaign.

ThumbGate is not a GitHub star package, not fake engagement, and not a substitute for npm installs or merged PRs. Real engagement is npx thumbgate init and a PreToolUse hook that actually fires.

Tech memes (shareable)

Lightweight visuals for how agents fail without a pre-action gate:

It hard-blocks detected secret leaks and two direct self-disable command classes by default — commands that terminate the ThumbGate gate process or enable its bypass environment override. Other high-risk classes (rm -rf, force-push, fetch-and-run, direct guardrail edits) warn and log by default. Set THUMBGATE_STRICT_ENFORCEMENT=1 for strict enforcement (warnings become hard denies).

Accepted feedback is stored as local lessons. Repeated concrete failures can become prevention rules that promote from warnings to blocking gates. The firewall improves from operations without retraining the model. Prompt evaluation (npx thumbgate eval) turns accepted feedback into reusable eval cases and local proof reports.

Honest disclaimer: ThumbGate does not update model weights. It intercepts tool calls at runtime. Local-first — no cloud required for the enforcement path.

Works with Claude Code, Cursor, Codex, Gemini CLI, Amp, Cline, OpenCode, and other MCP agents.

  Agent tries:   rm -rf tests/

ThumbGate: 👎 WARN + LOG — "Never delete test directories" Pattern matched: rm.*-rf.*tests Source: your thumbs-down from last Tuesday Strict mode: ⛔ DENY before tool execution

Agentic development cycle fit

Agentic development is becoming a loop: Guide → Generate → Verify → Solve. ThumbGate is the pre-action gate / pre-action boundary between generated intent and executed action.

---

Quick Start

Want a phased walkthrough with a verify step at every stage? Follow the Progressive Setup Guide.

Progressive wiring — prove the pipe before you turn matching on. Empty dashboard is success.

npx thumbgate init          # Phase 1: hooks only

npx thumbgate doctor # verify: exits 0 only when PreToolUse hook is wired (hidden metric = hook install, not gate count) npx thumbgate dashboard --open # Phase 2: open local HTML; empty stats are OK npx thumbgate capture --feedback=down --context="Never run DROP on production tables" --what-went-wrong="agent proposed DROP" --what-to-change="require review for DROP"

Later DROP attempts in the same scope surface the check:

⚠️ Check fired: "Never run DROP on production tables"

Pattern: DROP.*production Verdict: 👎 WARN + LOG (⛔ BLOCK when THUMBGATE_STRICT_ENFORCEMENT=1)

Numbered configs: config/progressive/. Guide: progressive wiring.

MCP / Glama / registry install (stdio)

Directories and clients that install ThumbGate as an MCP server must start stdio MCP, not the HTTP API:

npx -y thumbgate serve

• Equivalent: npx -y thumbgate mcp
• Do not use npm start for MCP — that launches the hosted HTTP API (src/api/server.js), not the agent-facing stdio server.

▶ 90-second demo · GIF walkthrough

---

Install for your agent

Per-agent guides: Claude/Codex bridge · Codex profile · Cursor · MCP setup

Install scope: machine-wide vs per-project

Both scopes write mcpServers.thumbgate plus PreToolUse / UserPromptSubmit / PostToolUse / SessionStart hooks. Machine-wide is the right default for most developers. Cross-repo blocking is not automatic: a lesson learned in one project only applies elsewhere when you share the store (machine-wide) or export/import lessons.

MCP tools (surface): gate_check (read/evaluate proposed tool call), feedback capture + session tools (write), dashboard/stats (read). Destructive agent actions stay blocked/warned by PreToolUse — ThumbGate does not execute user shell commands for you.

---

Discoverable slash-commands — the guardrail layer for spec-driven agents

Spec-driven agent frameworks like GSD (get-shit-done) and GitHub Spec Kit plan and generate work. ThumbGate is the guardrail layer for spec-driven agents: it sits *after* the plan, on the boundary between a generated tool call and its execution — alongside GSD / Spec-Kit, not instead of them.

npx thumbgate init installs these into your agent palette:

---

Pricing & buyer paths

Free tier: 2 feedback captures/day (10 total) and up to 3 active auto-promoted prevention rules. Pro ($19/mo or $149/yr) is the individual tier for unlimited rules, history-aware lessons, linked feedback session flow, personal dashboard, and DPO export. Enterprise is custom and scoped after intake; hosted team lesson sync and a hosted org dashboard are not general availability.

| Feedback captures | 2 feedback captures/day (10 total) | Unlimited | S