stevesolun/ctx / docs
skill-router
FlaggedRepo-aware recommendation manager for ctx. Scans the active repository, identifies stack and workflow signals, recommends a capped set of skills, agents, and MCP servers, and unloads helpers that no longer match the current work after user confirmation. Harnesses are recommended by the custom-model onboarding flow or loop adapters and then attach to the same recommendation layer.
huggingface-publish.md
View on GitHub# Hugging Face Publish
ctx publishes the GitHub repository as the public Hugging Face dataset repo
[`Stevesolun/ctx`](https://huggingface.co/datasets/Stevesolun/ctx). The
dataset repo is a clean `git ls-files` snapshot plus the two exact archives
declared by `graph/release-artifacts.json`, not local review reports or ignored
caches.
## What gets uploaded
- Tracked source, docs, tests, and packaging files.
- `graph/wiki-graph.tar.gz`.
- `graph/wiki-graph-runtime.tar.gz`.
- The compressed skill index under `graph/`.
- Tracked graph visualizations under `graph/`.
Ignored local reports, review notes, raw ingest caches, coverage files,
`site/`, and `.pytest_cache/` are not uploaded because they are not tracked
by git.
## Automatic publish
Every push to `main` runs `.github/workflows/huggingface-sync.yml`. The job
checks out source, strictly validates `graph/release-artifacts.json`, and
hydrates only its two large archives from the named GitHub release. It verifies
all five graph assets by exact size and SHA-256, with no Git LFS fallback. It
then installs the sync dependencies and calls
`scripts/sync_huggingface.py`. It publishes only when the repository secret
`HF_TOKEN` is configured. On the canonical
`stevesolun/ctx` repository, a missing token is a hard failure so main cannot
silently drift from the dataset repo. On forks, a missing token still exits
successfully with a notice because forks are not trusted publishing sources.
When only repo-card inputs changed (`README.md`, `CHANGELOG.md`, or files under
`docs/`), the workflow uses card-only upload mode. Source, test, workflow, graph,
or packaging changes always run the full dataset sync so tracked files cannot
drift behind GitHub.
The sync script is still the contract: it exports the tracked git snapshot,
adds the two manifest-declared archives, adds Hugging Face repo-card metadata,
validates README/docs stats, and refuses missing, stale, or corrupt artifacts.
## Manual publish
Use the repository sync script. It exports tracked files plus the validated
local graph artifacts, adds the Hugging Face repo-card frontmatter to the
uploaded `README.md`, and refuses to publish if the manifest or any full wiki,
runtime wiki, or compressed skill-index byte identity is missing or mismatched.
Full sync uploads the exported tree with `delete_patterns="*"`, so files removed
from the current git snapshot are removed remotely in the same commit.
Card-only sync exports only `README.md`, `CHANGELOG.md`, and tracked `docs/**`
inputs, then replaces just those remote paths.
Do not paste the token into a command line. Prompt for it, set it only for the
current process, and clear it after the upload.
```bash
python -m pip install --upgrade huggingface_hub
read -rsp "HF write token: " HF_TOKEN
printf '\n'
export HF_TOKEN
trap 'unset HF_TOKEN' EXIT
python scripts/sync_huggingface.py --repo . --repo-id Stevesolun/ctx --repo-type dataset
unset HF_TOKEN
trap - EXIT
```
For a README/changelog/docs-only refresh:
```bash
python scripts/sync_huggingface.py --repo . --repo-id Stevesolun/ctx --repo-type dataset --card-only
```
## Verify
```bash
python - <<'PY'
from huggingface_hub import HfApi
api = HfApi()
info = api.repo_info(repo_id="Stevesolun/ctx", repo_type="dataset")
print(info.id, info.sha)
PY
```
The dataset page should show the MIT license and the tags from the metadata
wrapper.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |