AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
MCP unreviewed MIT Self-run

Arc Cli

mcp-ashutosh0x-arc-cli · by Ashutosh0x

ARC: High-Performance Agentic CLI Framework built in Rust for autonomous multi-agent code generation.

No reviews yet
0 installs
20 views
0.0% view→install

Install

$ agentstack add mcp-ashutosh0x-arc-cli

Open-source listing, not yet scanned by AgentStack. Follow the source repository for install instructions.

Security review

⚠ Flagged

1 finding(s); flagged for manual review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures
  • high Pipes remote content directly into a shell (remote code execution).

What it can access

  • Network access Used
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Reliability & compatibility

Not yet reviewed
0 installs to date
no reviews yet
5mo ago

Declared compatibility

Claude CodeClaude DesktopCursorWindsurf

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Arc Cli? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

ARC

[](https://github.com/Ashutosh0x/arc-cli/releases/latest) [](https://github.com/Ashutosh0x/arc-cli/releases) [](https://github.com/Ashutosh0x/arc-cli/actions) [](LICENSE)

[](https://www.rust-lang.org/) [](https://scorecard.dev/viewer/?uri=github.com/Ashutosh0x/arc-cli) [](https://codecov.io/gh/Ashutosh0x/arc-cli) [](https://crates.io/crates/arc-cli)

[](https://platform.openai.com/) [](https://www.anthropic.com/) [](https://groq.com/) [](https://x.ai/) [](https://ai.google.dev/) [](https://ollama.com/)

A native agentic CLI that reasons over codebases. One binary. No runtime. Works offline.


Install

# Linux / macOS
curl -fsSL https://raw.githubusercontent.com/Ashutosh0x/arc-cli/main/install.sh | sh

# Windows
irm https://raw.githubusercontent.com/Ashutosh0x/arc-cli/main/install.ps1 | iex

# From source
cargo install --git https://github.com/Ashutosh0x/arc-cli

What it does

ARC is a terminal-native AI coding agent. You point it at a codebase, describe what you want, and it plans, writes, and verifies the changes autonomously.

It connects to Claude, Gemini, OpenAI, Groq, xAI Grok, or Ollama (fully offline). You pick the provider. Switch live with /provider groq. If one goes down, it fails over to the next.

arc chat                    # start a session
/plan "add OAuth2 login"    # generate a plan, review it, accept or reject
/provider groq              # switch to Groq (Llama 3.3 70B)
/model grok-4-1-fast-non-reasoning  # switch model
/checkpoint                 # snapshot current state
/rewind 3                   # undo to checkpoint 3
/compact                    # compress context window
arc review                  # PR critique via 6 specialized agents

Why not Claude Code or Gemini CLI?

| | ARC | Claude Code | Gemini CLI | |---|-----|-------------|------------| | Written in | Rust | Node.js | Node.js | | Cold start | | Structural search: functions, classes, types | | arc graph trace | Call graph — who calls what, BFS depth 1-5 | | arc graph architecture | Architecture overview: layers, clusters, hotspots | | arc graph impact | Map git diff → affected symbols + risk scores | | arc graph query | Execute Cypher-like graph queries | | /plan [task] | Generate and review a modification plan | | /provider [name] | Switch provider live (anthropic, groq, xai, openai) | | /model [name] | Switch model (e.g. grok-4-1-fast-non-reasoning) | | /checkpoint | Save session state | | /rewind [id] | Restore a previous checkpoint | | /compact | Compress context window | | /status | Show current provider, model, message count | | /memory [k] [v] | Persistent key-value store | | /fork [name] | Branch the conversation | | /security-review | Scan diffs for vulnerability patterns | | /copy` | Pick and copy code blocks |

Code Intelligence

ARC integrates with codebase-memory-mcp to provide structural code intelligence via a persistent knowledge graph.

arc graph index                                    # index your project (seconds for typical repos)
arc graph search ".*Handler.*"                     # find all handler functions
arc graph trace authenticate --direction inbound   # who calls authenticate()?
arc graph architecture                             # languages, packages, routes, hotspots
arc graph impact                                   # what breaks from your uncommitted changes?
arc graph query 'MATCH (f:Function)-[:CALLS]->(g) WHERE f.name = "main" RETURN g.name'

| Metric | Without graph | With graph | |--------|--------------|------------| | Context tokens per query | ~412,000 | ~3,400 | | Symbol lookup | 50-500ms (grep) | <1ms | | Languages parsed | 5 | 64 | | Call graph | No | Yes, BFS depth 1-5 | | Dead code detection | No | Yes, ~150ms | | Session memory | Resets | Persists (SQLite) |

Diff review keybindings

When ARC proposes file changes, you review them interactively:

| Key | Action | |-----|--------| | y / Enter | Accept this change | | n / Esc | Reject this change | | a | Accept all remaining | | d | Reject all remaining | | e | Open in $EDITOR | | j / k | Scroll through large diffs |

Docs

  • [Getting Started](docs/getting-started.md)
  • [Architecture](docs/architecture.md)
  • [Security Model](docs/authentication.md)
  • [Provider Setup](docs/providers.md)
  • [MCP Integration](docs/mcp.md)
  • [Benchmarks](docs/benchmarks.md)

Contributing

See [CONTRIBUTING.md](CONTRIBUTING.md). The codebase enforces #![forbid(unsafe_code)], cargo fmt, and cargo clippy on every PR.

License

MIT

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.