corrosive_agents 0.0.1

Build verifiable, interactive AI agents powered by NVIDIA Nemotron free LLM models — MCP, skills, Ed25519 identity, REST/WebSocket/gRPC transports, and pluggable vector stores (Pinecone, Qdrant, custom).
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138

#![warn(missing_docs)]
#![cfg_attr(docsrs, feature(doc_cfg))]

//! # corrosive_agents
//!
//! Build **verifiable, interactive AI agents** powered by
//! [NVIDIA Nemotron](https://build.nvidia.com) free LLM models.
//!
//! ## Highlights
//!
//! - **Builder pattern** — assemble an [`Agent`] fluently with [`AgentBuilder`].
//! - **JSON manifests** — an agent's name, version, capabilities, skills and
//!   MCP servers load from a JSON file ([`AgentManifest`]).
//! - **Verifiable identity** — Ed25519 signatures over the manifest
//!   ([`identity::AgentIdentity`]); anyone can verify an agent with its
//!   public key.
//! - **NVIDIA NIM client** — chat + streaming + embeddings against the
//!   OpenAI-compatible `integrate.api.nvidia.com` endpoint
//!   ([`llm::NvidiaClient`]).
//! - **Skills & MCP** — native async [`skills::Skill`]s plus a
//!   [Model Context Protocol](https://modelcontextprotocol.io) stdio client.
//! - **Multi-transport** — Tokio REST API + WebSocket (feature `server`),
//!   and gRPC (feature `grpc`).
//! - **Vector stores** — one [`vector::VectorStore`] trait; Pinecone
//!   (feature `pinecone`), Qdrant (feature `qdrant`), an in-memory reference
//!   implementation, or your own custom store.
//!
//! ## Quickstart
//!
//! ```no_run
//! use corrosive_agents::prelude::*;
//!
//! # async fn run() -> corrosive_agents::Result<()> {
//! let agent = Agent::builder()
//!     .name("research-agent")
//!     .version("0.1.0")
//!     .description("Answers research questions")
//!     .capability(Capability::new("chat", "Conversational Q&A"))
//!     .llm(NvidiaClient::from_env()?.with_model(models::LLAMA_NEMOTRON_SUPER_49B))
//!     .generate_identity()
//!     .build()?;
//!
//! let reply = agent.chat("session-1", "What is Rust?").await?;
//! println!("{reply}");
//! # Ok(())
//! # }
//! ```
//!
//! ## Feature flags
//!
//! | Feature    | Default | Enables                                   |
//! |------------|---------|-------------------------------------------|
//! | `server`   | yes     | REST + WebSocket serving (axum)           |
//! | `grpc`     | no      | gRPC serving (tonic, vendored protos)     |
//! | `pinecone` | no      | Pinecone vector store backend             |
//! | `qdrant`   | no      | Qdrant vector store backend               |
//! | `full`     | no      | All of the above                          |

pub mod a2a;
pub mod agent;
pub mod error;
pub mod identity;
pub mod llm;
pub mod mcp;
pub mod session;
pub mod skills;
pub mod trust;
pub mod vector;

#[cfg(any(feature = "server", feature = "grpc"))]
#[cfg_attr(docsrs, doc(cfg(any(feature = "server", feature = "grpc"))))]
pub mod auth;

#[cfg(feature = "server")]
#[cfg_attr(docsrs, doc(cfg(feature = "server")))]
pub mod server;

#[cfg(feature = "grpc")]
#[cfg_attr(docsrs, doc(cfg(feature = "grpc")))]
pub mod grpc;

#[cfg(feature = "tls")]
#[cfg_attr(docsrs, doc(cfg(feature = "tls")))]
pub mod tls;

#[cfg(feature = "x509")]
#[cfg_attr(docsrs, doc(cfg(feature = "x509")))]
pub mod x509;

/// The step-by-step tutorial: from an empty project to a production-shaped
/// agent. Rendered from `docs/TUTORIAL.md`; its core code blocks are
/// compile-checked as doctests.
pub mod tutorial {
    #![doc = include_str!("../docs/TUTORIAL.md")]
}

pub use agent::{Agent, AgentBuilder, AgentInfo, AgentManifest, Capability};
pub use error::{Error, Result};

/// Convenient glob-import for the most commonly used types.
///
/// ```
/// use corrosive_agents::prelude::*;
/// ```
pub mod prelude {
    pub use crate::a2a::RemoteAgent;
    pub use crate::agent::{Agent, AgentBuilder, AgentInfo, AgentManifest, Capability};
    pub use crate::error::{Error, Result};
    pub use crate::identity::AgentIdentity;
    pub use crate::llm::{
        models, ChatMessage, ChatRequest, ChatResponse, EmbeddingProvider, LlmProvider,
        NvidiaClient, RetryPolicy, StreamChunk, ToolCall, ToolSpec, UsageEvent, UsageObserver,
        UsageSnapshot,
    };
    pub use crate::mcp::{McpClient, McpPrompt, McpResource, McpServerConfig, McpTool};
    pub use crate::session::{InMemorySessionStore, SessionStore};
    pub use crate::skills::{FnSkill, Skill, SkillPolicy, SkillRegistry};
    pub use crate::trust::{Revocation, RotationProof, TrustStore};
    pub use crate::vector::{
        chunk_text, Document, InMemoryVectorStore, MetadataFilter, SearchResult, VectorStore,
    };

    #[cfg(any(feature = "server", feature = "grpc"))]
    pub use crate::auth::{AuthScheme, JwksStore};
    #[cfg(feature = "tls")]
    pub use crate::tls::TlsConfig;
    #[cfg(feature = "pgvector")]
    pub use crate::vector::PgVectorStore;

    #[cfg(feature = "redis-sessions")]
    pub use crate::session::RedisSessionStore;
    #[cfg(feature = "sqlite-sessions")]
    pub use crate::session::SqliteSessionStore;
    #[cfg(feature = "pinecone")]
    pub use crate::vector::PineconeStore;
    #[cfg(feature = "qdrant")]
    pub use crate::vector::QdrantStore;
}