You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
docs: add TL;DR and progressive disclosure to cookbook page (category 1)
Collapse helper script details behind expandable blocks so the page
is faster to scan. Add TL;DR with anchor links to all 8 recipes.
Co-authored-by: Ralphify <noreply@ralphify.co>
8 copy-pasteable ralph loops: [autoresearch](#autoresearch), [codebase improvement](#codebase-improvement), [documentation](#documentation-loop), [bug hunting](#bug-hunter), [deep research](#deep-research), [code migration](#code-migration), [security scanning](#security-scan), and [test coverage](#test-coverage). Each is a real, runnable example from the `examples/` directory.
11
+
9
12
Copy-pasteable setups for common autonomous workflows. Each recipe is a real, runnable ralph from the [`examples/`](https://github.com/computerlovetech/ralphify/tree/main/examples) directory.
10
13
11
14
All recipes use **Claude Code** as the agent. To use a different agent, swap the `agent` field — see [Using with Different Agents](agents.md).
@@ -25,19 +28,19 @@ An autonomous ML research loop inspired by [Karpathy's autoresearch](https://git
25
28
--8<-- "examples/autoresearch/RALPH.md"
26
29
```
27
30
28
-
The helper scripts surface experiment state each iteration:
31
+
??? note "Helper scripts that surface experiment state each iteration"
29
32
30
-
**`autoresearch/show-results.sh`**
33
+
**`autoresearch/show-results.sh`**
31
34
32
-
```bash
33
-
--8<-- "examples/autoresearch/show-results.sh"
34
-
```
35
+
```bash
36
+
--8<-- "examples/autoresearch/show-results.sh"
37
+
```
35
38
36
-
**`autoresearch/show-last-run.sh`**
39
+
**`autoresearch/show-last-run.sh`**
37
40
38
-
```bash
39
-
--8<-- "examples/autoresearch/show-last-run.sh"
40
-
```
41
+
```bash
42
+
--8<-- "examples/autoresearch/show-last-run.sh"
43
+
```
41
44
42
45
```bash
43
46
ralph run autoresearch --train_script train.py --prepare_script prepare.py
@@ -155,7 +158,8 @@ This is a more advanced ralph — it uses `args` for the research topic, helper
155
158
--8<-- "examples/research/RALPH.md"
156
159
```
157
160
158
-
The helper scripts (`show-focus.sh`, `show-questions.sh`, etc.) read from the workspace files and surface key state. The `review.sh` script pipes the full workspace to a separate Claude call that acts as an editorial reviewer — giving the research agent targeted feedback each iteration.
The helper scripts read from the workspace files and surface key state. The `review.sh` script pipes the full workspace to a separate Claude call that acts as an editorial reviewer — giving the research agent targeted feedback each iteration.
159
163
160
164
```bash
161
165
ralph run research --workspace ai-safety --focus "current approaches to AI alignment"
@@ -184,11 +188,12 @@ A loop for batch code transformations — migrating from one pattern to another
184
188
--8<-- "examples/migrate/RALPH.md"
185
189
```
186
190
187
-
The `count-remaining.sh` script receives the pattern as an argument (resolved from `{{ args.old_pattern }}` in the `run` field) to find files that still need migration:
0 commit comments