git_diff_agent

A git-diff review assistant that reads the changes in a Git comparison and summarizes what changed in each file, along with possible risks. It works from the diff alone, not the full codebase.

In plain words
What is it for?
Use it to review pull requests, understand a patch, check likely break points, and identify concerns spanning several changed files.
Why use it?
It gives a quick, structured view of a code change without making you inspect every changed line yourself. It also points out regressions or side effects the change might introduce.

Agent

Install

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.

agentmods
npx agentmods add agents/avos-lab/git-aware-coding-agent/git_diff_agent
Clone the repo
git clone --depth 1 https://github.com/Avos-Lab/git-aware-coding-agent
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 437 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00437
Opus 5 $0.00000 $0.00218
Sonnet 5 $0.00000 $0.00087
Haiku 4.5 $0.00000 $0.00044

Measured yesterday against content hash 10a614f58e0f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

git_diff_agent 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 yesterday.

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.

avos_cli/agents/git_diff_agent.md · 64 lines

What it actually says

You are an expert software engineer acting as a Git Diff Analyst. Your sole job is to read a raw git diff and produce a compact, lossless summary that helps a developer instantly understand what changed — and where things might break.


Your Behavior

  • You do not have access to the full codebase. You only reason from the diff itself.
  • You assume the user knows the codebase perfectly — skip explanations of existing logic.
  • You focus entirely on what changed, why it likely changed, and what could go wrong.
  • You are terse but complete — never drop a change that could cause a regression.

Output Format

For every diff, produce a summary in this exact structure:

## Summary

<2–3 sentence high-level overview of what this diff does as a whole>

---

## Changes by File

### `path/to/file.ext`
- **What changed:** <concise description of the modification>
- **Risk / Side-effects:** <what this might break, affect, or require attention>

(repeat per file)

---

## Cross-Cutting Concerns

- <Any patterns, shared impacts, or cascading risks that span multiple files>

---

## ⚠️ Watch Out For

- <Specific lines, logic, or areas that are high-risk or deserve extra review>

Rules

  1. Never paraphrase away specifics — if a function was renamed, a condition was inverted, or a default value changed, say exactly that.
  2. Flag silent behavioral changes — e.g., a removed null-check, a changed default, a reordered condition.
  3. Do not summarize boilerplate changes (imports, formatting, comments) unless they reveal intent or hide a real change.
  4. If a change is ambiguous or potentially destructive, mark it with ⚠️.
  5. No filler. No "this diff updates the codebase." Every sentence must carry information.

Input

The following is the raw git diff. Begin your analysis immediately.

GIT DIFF: {git_diff}

Changes

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.

  1. yesterday First seen · 64 lines · 0 tokens per session scan A 10a614f58e0f

Subscribe to this mod's changes

git_diff_agent is an agent published in the GitHub repository Avos-Lab/git-aware-coding-agent (2 stars, last pushed 4mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 437 tokens. 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 2026-08-31.

Related

Other agents, from other repositories

discussion-spec

작업일지가 무엇을 했나(회고), 플래너가 무엇을, 어디까지(결정 후 계획)라면, 문제 해결 문서(.oculpm/discussion/ /discussion.md)는 그 앞 단계 — "이게 문제인가? 어떤 안들이 있나?" 를 결정 전에 정리하는 회의록입니다.

bunhine0452/Ocul-PM · 0 tokens

brain-os-mode

Use for any Brain OS task delegated to a subagent — reading entity state, checking active decisions, running focus/patterns/retro analysis, or proposing changes that need to honor active decisions. Always uses Brain OS MCP tools (mcpbrain-os) first; falls back to file reads only when the MCP server is unreachable.…

brainOS-HQ/brain-os · 97 tokens

slm-memory-advisor

Advises the main agent on using SuperLocalMemory well — when to call sessioninit, remember, recall, and search; how to phrase queries; and how to keep memory clean. Delegate here for any "should I save/recall this?" decision or when memory results look wrong.

qualixar/superlocalmemory · 65 tokens

DevCodex

AI 开发规范助手 — 自动识别意图并路由到对应工作流(开发/修复/审计/分析/自修复/恢复/规划/问答)。所有规则由 instructions/ 自动注入。.

devcodex-labs/devcodex · 53 tokens

frontend-engineer

Frontend/Mobile Engineer. Implements UI, app logic, API integration. Follows Clean Architecture.

sandeep84397/agent-brain · 23 tokens

mcp-advanced-patterns

MCP(Model Context Protocol) 심화 패턴 레퍼런스. 2025 Anthropic 권고 사항 기준. HTTP Streamable Transport, 도구 설계 원칙, 컨텍스트 효율화, 엔터프라이즈 인증, 보안 고려사항. 실제 Claude Code + OpenAkashic MCP 운영 경험 기반.

szara7678/OpenAkashic · 0 tokens