Files
everything-claude-code/skills/gateguard/SKILL.md
seto 4dbed5ff5b fix: cubic-dev-ai round 2 — 3 issues across SKILL.md + pruning
P1: Gate message asked for raw production data records — changed to
    "redacted or synthetic values" to prevent sensitive data exfiltration

P2: SKILL.md description now includes MultiEdit (was missing after
    MultiEdit gate was added in previous commit)

P2: Session key pruning now caps __prefixed keys at 50 to prevent
    unbounded growth even in theoretical edge cases

9/9 tests pass.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-13 16:11:33 +09:00

4.2 KiB

name, description, origin
name description origin
gateguard Fact-forcing gate that blocks Edit/MultiEdit/Write/Bash and demands concrete investigation (importers, data schemas, user instruction) before allowing the action. Measurably improves output quality by +2.25 points vs ungated agents. community

GateGuard — Fact-Forcing Pre-Action Gate

A PreToolUse hook that forces Claude to investigate before editing. Instead of self-evaluation ("are you sure?"), it demands concrete facts. The act of investigation creates awareness that self-evaluation never did.

When to Activate

  • Working on any codebase where file edits affect multiple modules
  • Projects with data files that have specific schemas or date formats
  • Teams where AI-generated code must match existing patterns
  • Any workflow where Claude tends to guess instead of investigating

Core Concept

LLM self-evaluation doesn't work. Ask "did you violate any policies?" and the answer is always "no." This is verified experimentally.

But asking "list every file that imports this module" forces the LLM to run Grep and Read. The investigation itself creates context that changes the output.

Three-stage gate:

1. DENY  — block the first Edit/Write/Bash attempt
2. FORCE — tell the model exactly which facts to gather
3. ALLOW — permit retry after facts are presented

No competitor does all three. Most stop at deny.

Evidence

Two independent A/B tests, identical agents, same task:

Task Gated Ungated Gap
Analytics module 8.0/10 6.5/10 +1.5
Webhook validator 10.0/10 7.0/10 +3.0
Average 9.0 6.75 +2.25

Both agents produce code that runs and passes tests. The difference is design depth.

Gate Types

Edit Gate (first edit per file)

Before editing {file_path}, present these facts:

1. List ALL files that import/require this file (use Grep)
2. List the public functions/classes affected by this change
3. If this file reads/writes data files, show field names, structure,
   and date format (use redacted or synthetic values, not raw production data)
4. Quote the user's current instruction verbatim

Write Gate (first new file creation)

Before creating {file_path}, present these facts:

1. Name the file(s) and line(s) that will call this new file
2. Confirm no existing file serves the same purpose (use Glob)
3. If this file reads/writes data files, cat one real record
4. Quote the user's current instruction verbatim

Destructive Bash Gate (every destructive command)

Triggers on: rm -rf, git reset --hard, git push --force, drop table, etc.

1. List all files/data this command will modify or delete
2. Write a one-line rollback procedure
3. Quote the user's current instruction verbatim

Routine Bash Gate (once per session)

Quote the user's current instruction verbatim.

Quick Start

Option A: Use the ECC hook (zero install)

The hook at scripts/hooks/gateguard-fact-force.js is included in this plugin. Enable it via hooks.json.

Option B: Full package with config

pip install gateguard-ai
gateguard init

This adds .gateguard.yml for per-project configuration (custom messages, ignore paths, gate toggles).

Anti-Patterns

  • Don't use self-evaluation instead. "Are you sure?" always gets "yes." This is experimentally verified.
  • Don't skip the data schema check. Both A/B test agents assumed ISO-8601 dates when real data used %Y/%m/%d %H:%M. Checking one real record prevents this entire class of bugs.
  • Don't gate every single Bash command. Routine bash gates once per session. Destructive bash gates every time. This balance avoids slowdown while catching real risks.

Best Practices

  • Let the gate fire naturally. Don't try to pre-answer the gate questions — the investigation itself is what improves quality.
  • Customize gate messages for your domain. If your project has specific conventions, add them to the gate prompts.
  • Use .gateguard.yml to ignore paths like .venv/, node_modules/, .git/.
  • safety-guard — Runtime safety checks (complementary, not overlapping)
  • code-reviewer — Post-edit review (GateGuard is pre-edit investigation)