Skip to content

knewstimek/agent-tool

v0.7.10 Breaking

This release includes breaking changes for platform teams planning a safe upgrade.

Published 2mo MCP Developer Tools
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

ai-coding automation claude-code cli codex cross-platform
+13 more
cursor developer-tools dns encoding file-tools go mcp mcp-server mysql redis sftp ssh tls

Summary

AI summary

Analyze now supports ELF and Mach‑O binaries, call graph across multiple architectures, and removed global size limits.

Full changelog

Quick Start

  1. Download the binary for your OS below
  2. Run agent-tool install (or agent-tool install claude)
  3. Restart your IDE
  4. Done -- all tools work immediately, no permission popups

Or just ask your AI agent:

"Download agent-tool from https://github.com/knewstimek/agent-tool/releases/latest and run agent-tool install"

Any capable AI coding agent (Claude Code, Codex, etc.) can handle the full download -> install -> restart flow automatically.

Tip: Add this to your CLAUDE.md or AGENTS.md so your agent prefers agent-tool over built-in tools:

Strict mode:

ALWAYS use agent-tool MCP tools (mcp__agent-tool__*) instead of built-in file tools. agent-tool preserves file encoding and respects .editorconfig indentation settings.

Soft mode:

Prefer agent-tool MCP tools (mcp__agent-tool__*) over built-in file tools when available.


What's New in v0.7.10

analyze: Multi-format binary analysis

  • disassemble: ELF and Mach-O support with VA resolution and arch auto-detect (previously PE only)
  • call_graph: PE/ELF/Mach-O support for x86, x64, ARM64, and ARM32 architectures
    • ARM64: BL imm26 target-based heuristic function detection
    • ARM32: BL imm24 with PC+8 pipeline offset
  • No file size limit: Removed global size cap for analyze operations. Multi-GB binaries now work via lazy section-based reading. Only strings and entropy (which load entire files) retain per-operation limits.

read: Flexible offset

  • offset now accepts integer, string range "100-200", or [start, end] array
  • Agents no longer get validation errors when passing offset as a string or array

grep: Output modes and context lines

  • New output_mode parameter:
    • content (default): matching lines with path:line:text
    • files_with_matches: file paths only (fast, short-circuits on first match)
    • count: match count per file
  • New context parameters (like grep -C/-B/-A):
    • context: lines before and after each match
    • before / after: asymmetric context (overrides context)
    • Adjacent context regions are automatically merged with -- separators
  • Token savings: Single-file search omits the file path prefix from each line (agent already knows which file it passed). Directory search still includes path on every line.

Full Changelog: https://github.com/knewstimek/agent-tool/compare/v0.7.9...v0.7.10

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track knewstimek/agent-tool

Get notified when new releases ship.

Sign up free

About knewstimek/agent-tool

Encoding-aware, indentation-smart file tools for AI coding agents. 20+ tools including read/edit with automatic encoding detection, smart indentation conversion, SSH, SFTP, process management, and system utilities. Preserves file encoding (UTF-8, EUC-KR, Shift_JIS, etc.) and respects .editorconfig settings.

All releases →

Beta — feedback welcome: [email protected]