loopsmith-core 0.3.1

Config model and A-H validation for loopsmith loops
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

<div align="center">
  <img src="https://raw.githubusercontent.com/bitphill/loopsmith/v0.3.1/assets/loopsmith-logo-256.png" alt="loopsmith" width="140" />
  <h1>loopsmith-core</h1>
  <p><em>The A–J config model for loopsmith loops, and its validation.</em></p>
</div>

[![crates.io](https://img.shields.io/crates/v/loopsmith-core?logo=rust&logoColor=white&label=crates.io&color=e6522c)](https://crates.io/crates/loopsmith-core)
[![license](https://img.shields.io/badge/license-MIT-C8CAD1?labelColor=222)](https://github.com/bitphill/loopsmith/blob/main/LICENSE)
![rust](https://img.shields.io/badge/rust-1.75%2B-C1272D?logo=rust&logoColor=white)

Part of **[loopsmith](https://github.com/bitphill/loopsmith)** — self-evolving
agent loops behind a deterministic verification gate.

> **You probably do not need to depend on this directly.** It is a component of the
> `loopsmith` binary and compiles automatically as one of its dependencies:
>
> ```bash
> cargo install loopsmith
> ```
>
> Depend on it directly only if you are building something else on loopsmith's
> internals. The API is not yet stable across minor versions.

## What this crate is

A loop is described by ten sections, **A** to **J**: information, the manual
work list, goals, validations, success criteria, stop gates, schedules,
constraints, execution guidelines, default skills. This crate is that model,
its parser for both YAML and Markdown, and the validation that decides whether
a config is coherent.

The load-bearing behaviour: **validation fails while any `pre_execution` step is
unfinished.** Automating a process you cannot describe in checkable terms
produces fast, confident garbage, so a config that has not been run by hand once
is refused rather than accommodated.

## Where it sits

```
loopsmith  (the CLI binary)
└── loopsmith-mcp ── loopsmith-gate ─┐
    loopsmith-skills ────────────────┤
    loopsmith-provider ──────────────┼── loopsmith-core ── loopsmith-util
    loopsmith-graph ─────────────────┤        (config)      (primitives)
    loopsmith-memory ────────────────┘
```

| Crate | Purpose |
|---|---|
| [`loopsmith`](https://crates.io/crates/loopsmith) | the CLI binary |
| [`loopsmith-util`](https://crates.io/crates/loopsmith-util) | PATH lookup, wall clock, runtime platform detection |
| [`loopsmith-core`](https://crates.io/crates/loopsmith-core) | the A–J config model and its validation |
| [`loopsmith-memory`](https://crates.io/crates/loopsmith-memory) | `sled`-backed episodes, goal state, ledger, checkpoints |
| [`loopsmith-graph`](https://crates.io/crates/loopsmith-graph) | DAG scheduling, critical path, Amdahl-driven concurrency |
| [`loopsmith-gate`](https://crates.io/crates/loopsmith-gate) | the deterministic verification gate |
| [`loopsmith-provider`](https://crates.io/crates/loopsmith-provider) | provider routing and the tier cascade |
| [`loopsmith-skills`](https://crates.io/crates/loopsmith-skills) | sub-agent acquisition, quarantine, outcome ranking |
| [`loopsmith-mcp`](https://crates.io/crates/loopsmith-mcp) | local stdio MCP server over memory, gate, and graph |

## The one rule the whole design rests on

> A model must not be the thing that certifies its own completion.

`goal_satisfied` is written by [`loopsmith-gate`](https://crates.io/crates/loopsmith-gate)
and by nothing else, and the gate can **revoke**: delete a required artifact and a
satisfied goal flips back.

Full documentation: <https://github.com/bitphill/loopsmith#readme>

MIT licensed. © bitphill