원클릭으로 Manus에서 모든 스킬 실행

patch-trust-score

스타15

포크2

업데이트2026년 5월 30일 10:08

Assign a confidence and risk rating to every AI-generated change. Returns safe, review-required, or high-risk verdict with specific signals.

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

DVNghiem

DVNghiem/FlowDeck

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

Patch Trust Score

Every AI-generated patch gets a trust score before it is applied. The score gates how the change is handled: auto-apply, flag for review, or block and escalate.

Score Bands

Score	Verdict	Action
80–100	safe	Auto-apply in guarded mode
40–79	review-required	Inline warning + human ACK
0–39	high-risk	Block auto-apply, route to reviewer

Risk Signals (each reduces the score)

Signal	Deduction
File has prior failure history	−20
Edit contains auth/crypto/payment keywords	−8 per keyword (max −30)
File in arch-constrained zone	−20
No test coverage for file	−10

Workflow

For every proposed write or edit: a. Check .codebase/FAILURES.json for prior failures on this file b. Scan the patch content for high-risk keywords c. Check .codebase/CONSTRAINTS.md for boundary violations
Compute score (0–100, start at 100)
Emit verdict with signals
Route accordingly (auto / warn / block)

Integration

The patch-trust hook runs automatically on every write and edit tool call. The score is logged to stdout and appended to .codebase/DECISIONS.jsonl.

For manual scoring, use the patch-trust-score skill with a file path and change description.

이 저장소의 다른 Skills

같은 저장소

context-budget

DVNghiem/FlowDeck

Optimize token usage and context window discipline. Reduce costs and improve response quality through smart context management.

2026-07-2215

context-steward

DVNghiem/FlowDeck

Unified context lifecycle for FlowDeck sessions — ingest, filter, prune, protect, summarize, and persist with telemetry.

2026-07-2215

change-impact-radar

DVNghiem/FlowDeck

Predict affected files, modules, APIs, tests, and DB paths before changes. Returns an impact map for human review.

2026-07-2215

codebase-mapping

DVNghiem/FlowDeck

Map architecture, conventions, and file structure into `.codebase/`. Use when onboarding or before deep feature work.

2026-07-2215

confidence-aware-planning

DVNghiem/FlowDeck

Plan differently when the agent has low certainty — ask for clarification or narrow scope instead of pretending full understanding.

2026-07-2215

context-guard

DVNghiem/FlowDeck

Protect critical context from pruning during compaction. Preserve active plans, safety files, pending operations, and user intent anchors.

2026-07-2215

name	patch-trust-score
description	Assign a confidence and risk rating to every AI-generated change. Returns safe, review-required, or high-risk verdict with specific signals.
origin	FlowDeck

Patch Trust Score

Every AI-generated patch gets a trust score before it is applied. The score gates how the change is handled: auto-apply, flag for review, or block and escalate.

Score Bands

Score	Verdict	Action
80–100	safe	Auto-apply in guarded mode
40–79	review-required	Inline warning + human ACK
0–39	high-risk	Block auto-apply, route to reviewer

Risk Signals (each reduces the score)

Signal	Deduction
File has prior failure history	−20
Edit contains auth/crypto/payment keywords	−8 per keyword (max −30)
File in arch-constrained zone	−20
No test coverage for file	−10

Workflow

For every proposed write or edit: a. Check .codebase/FAILURES.json for prior failures on this file b. Scan the patch content for high-risk keywords c. Check .codebase/CONSTRAINTS.md for boundary violations
Compute score (0–100, start at 100)
Emit verdict with signals
Route accordingly (auto / warn / block)

Integration

The patch-trust hook runs automatically on every write and edit tool call. The score is logged to stdout and appended to .codebase/DECISIONS.jsonl.

For manual scoring, use the patch-trust-score skill with a file path and change description.