Skip to content

docs(skills): normalize profiling skill for Codex #64

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Jammy2211 merged 2 commits into from
Jul 11, 2026
Merged

docs(skills): normalize profiling skill for Codex #64

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5cd1937
docs(skills): normalize profiling skill for Codex
Jul 11, 2026
7a53eb0
fix(skills): route profiling setup through start-dev
Jul 11, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 7 additions & 5 deletions skills/profile_likelihood/SKILL.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,6 @@
---
name: profile_likelihood
name: profile-likelihood
description: Run a multi-config JAX likelihood profiling sweep (CPU/GPU × fp64/mp, optional A100 on RAL HPC) for a given likelihood function. Produces structured JSON+PNG outputs with comparison tables under autolens_workspace_developer/jax_profiling/results/jit/.
user-invocable: true
---

# Profile Likelihood
Expand All @@ -18,14 +17,17 @@ workflow. Verbose command blocks, gotchas and run precedent are in
## Usage

```
/profile_likelihood <dataset_type> <likelihood_type>
$profile-likelihood <dataset_type> <likelihood_type>
```

In Claude, invoke the same skill as `/profile_likelihood`.

`<dataset_type>` ∈ `imaging` / `interferometer` / `point_source`.
`<likelihood_type>` is a canonical reference script present at
`autolens_workspace_developer/jax_profiling/jit/<dataset_type>/<likelihood_type>.py`
(mge, delaunay, rectangular, delaunay_mge, rectangular_mge, …). Examples:
`/profile_likelihood imaging mge`, `/profile_likelihood interferometer rectangular_mge`.
`$profile-likelihood imaging mge`,
`$profile-likelihood interferometer rectangular_mge`.

## What this produces

Expand Down Expand Up @@ -54,7 +56,7 @@ GPU/HPC availability, the PyAutoGPU venv ordering, and the canonical reference).
only) — [`reference.md`](reference.md) → "Scaffold".
3. **SLURM submit scripts** (first time only) — [`reference.md`](reference.md) →
"SLURM submit scripts".
4. **Plan + worktree** — run `/plan_branches`; worktree
4. **Plan + worktree** — run `$start-dev` (`/start_dev` in Claude); worktree
`autolens_workspace_developer` on `feature/<likelihood_type>-profiling-a100`.
5. **Local sweep** — 4 configs (GPU/CPU × fp64/mp); spot-check `device.backend`.
6. **HPC sweep** — A100 fp64 + mp via `hpc/sync`, then consolidate.
Expand Down
7 changes: 4 additions & 3 deletions skills/profile_likelihood/reference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,9 +40,10 @@ both.

### Plan + worktree

Run `/plan_branches`. Affected: `z_projects/profiling` (no remote, local-only —
commit to local main) and `autolens_workspace_developer` (artifacts, has remote
— feature branch + PR). Branch `feature/<likelihood_type>-profiling-a100`.
Run `$start-dev` (`/start_dev` in Claude). Affected:
`z_projects/profiling` (no remote, local-only — commit to local main) and
`autolens_workspace_developer` (artifacts, has remote — feature branch + PR).
Branch `feature/<likelihood_type>-profiling-a100`.

```bash
source admin_jammy/software/worktree.sh
Expand Down
Loading