Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add manishiitg/coding-agent-loop --skill backupgit clone --depth 1 https://github.com/manishiitg/coding-agent-loopWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/manishiitg/coding-agent-loop/backup)<a href="https://agentmods.dev/skills/manishiitg/coding-agent-loop/backup"><img src="https://agentmods.dev/badge/skills/manishiitg/coding-agent-loop/backup/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/manishiitg/coding-agent-loop/backup"><img src="https://agentmods.dev/badge/skills/manishiitg/coding-agent-loop/backup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00034 | $0.00944 |
| Opus 5.5 | $0.00014 | $0.00378 |
| Sonnet 5.5 | $0.00007 | $0.00189 |
| Haiku 4.5 | $0.00003 | $0.00094 |
Grade A, and why
backup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Back up the workspace
Protect the family's data — materials, generated study/tests/reports, the progress
page, and conversations — by pushing it to the parent's OWN private Hugging Face
Hub dataset repo, via the hf CLI (already installed). This is the ONLY
destination this family uses — never suggest GitHub, Drive, S3, or anything else,
and never suggest a different destination "for images" — hf upload handles
binaries fine.
Same config vs status contract as publish/notify elsewhere in this app:
backup.json(workspace root) — declarative config:enabled,repo_id(the parent's HF username + a repo name, e.g.janedoe/sparkquill-backup).backup/status.json— the operational result of the last attempt:state(verified,pending, orfailed), last attempt/success timestamps, any error, and the source hash that was actually backed up. Never write status fields intobackup.json.
Steps
-
Check the token. The HF token lives in this app's own secrets store, not a bare
hf auth loginsession — calllist_secretsand look for one namedHF_TOKEN. If it's missing, tell the parent plainly: create a Write-access token at https://huggingface.co/settings/tokens, then save it viaset_secret(name it exactlyHF_TOKEN) or Settings → Secrets. Don't proceed without it. -
Read config.
cat backup.json. If missing, orenabledisn't true, orrepo_idis empty, this needs setup. -
First backup is ATTENDED — same rule as
publish. Never do the first (repo-creating) backup unattended: if this turn wasn't started by the parent directly asking (e.g. it's an automated Pulse check-in), do NOT ask questions into the void — just note ONCE that backup isn't set up yet ("ask your parent to say 'back up the workspace' to set it up"), don't repeat that note every cycle, and stop. If the parent IS present, ask for their HF username + a repo name (don't invent one), writebackup.jsonwithenabled: trueand thatrepo_id, then continue below. Oncebackup/status.jsonshowsstate: "verified"from a real prior success, later backups (including Pulse's own automated ones) may run without asking again.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed c9fcda9988c3
- 1mo ago First seen · 68 lines · 34 tokens per session scan A 8ba7cf599408
backup is a skill published in the GitHub repository manishiitg/coding-agent-loop (13 stars, last pushed 19d ago), licensed MIT. It adds 34 tokens to every session and 944 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
goga-cell-go
Golang rules for implementing CODEMANIFEST contracts.
configure-harness
Use when configuring which coding-agent harness octomux launches — per-task --model, harness selection, settings.harnesses. , resolveFlags/validateSettings, and registering an additional harness via a plugin.
data-agent-skill
Data processing and analysis specialist for the OpenClaw multi-agent system. Use this skill when the task involves: parsing CSV/JSON/Excel files, data cleaning and transformation, SQL queries, statistical analysis, generating charts or visualizations, aggregating data from multiple sources, or extracting insights from…
voice-ai-integration-engineer
Skill "voice-ai-integration-engineer" from aimerdoux/wavex-os, covering 🎙️ voice ai integration engineer agent, 🧠 your identity & memory, 🎯 your core mission, end-to-end transcription pipeline engineering and structured output and downstream integration.
model-qa
Skill "model-qa" from aimerdoux/wavex-os, covering model qa specialist, 🧠 your identity & memory, 🎯 your core mission, 1. documentation & governance review and 2. data reconstruction & quality.
data-engineer
Skill "data-engineer" from aimerdoux/wavex-os, covering data engineer agent, 🧠 your identity & memory, 🎯 your core mission, data pipeline engineering and data platform architecture.