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 big-emotion/agent-atelier --skill fix-traingit clone --depth 1 https://github.com/big-emotion/agent-atelierWrote 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/big-emotion/agent-atelier/fix-train)<a href="https://agentmods.dev/skills/big-emotion/agent-atelier/fix-train"><img src="https://agentmods.dev/badge/skills/big-emotion/agent-atelier/fix-train.svg" alt="Measured on agentmods" 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 | $0.00218 | $0.05767 |
| Opus 5 | $0.00109 | $0.02884 |
| Sonnet 5 | $0.00044 | $0.01153 |
| Haiku 4.5 | $0.00022 | $0.00577 |
Grade A, and why
fix-train 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 8d 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 — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fix Train
Walk a whole list of open pull requests and apply the changes that reviewers asked for — one pass, every PR. Collect each PR's requested changes, consolidate them into a concrete per-PR checklist, dispatch a team of parallel fixer sub-agents that implement the minimal fix and run the repo's quality gates, push to each PR branch, and report what was addressed and what still needs a human.
This is the third member of the train family and fills the gap between its siblings:
review-train— review only: posts reviews +approved/changes-requestedlabels. Writes nothing to code.fix-train— fix only (this skill): consumes the change requests, applies them, pushes. Never merges.merge-train— merge: orders, fixes CI/conflicts, merges.
The natural pipeline is review-train → fix-train → re-review → merge-train. fix-train is built to consume review-train's output directly (it reads the <!-- review-train --> "Issues requiring changes" list and the changes-requested label), but it works just as well from plain GitHub reviews left by humans.
This skill performs irreversible shared-repo actions (committing, pushing to PR branches, commenting). It follows the same grouped-confirmation model as its siblings: all collection and planning is read-only, one complete plan is presented, and it waits for a single explicit confirmation before any commit, push, or comment. Until the user confirms, nothing is written anywhere.
When to Activate
- User shares a GitHub pull-requests URL and asks to apply / address / fix the review feedback (not review, not merge).
- User says: "fix all the change requests", "address the review comments on every PR", "apply the requested changes across the queue", "clear the changes-requested PRs", "make the fixes the reviewers asked for".
- User invokes
/fix-train(optionally with a GitHub PRs URL or--dry-run).
If the user asks to review the PRs, defer to review-train. If they ask to merge / order / fix CI & conflicts then merge, defer to merge-train. Do not blur the three.
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.
- 8d ago First seen · 210 lines · 218 tokens per session scan A 3b0f702a3991
fix-train is a skill published in the GitHub repository big-emotion/agent-atelier (2 stars, last pushed today), licensed MIT. It adds 218 tokens to every session and 5,767 once invoked, about $0.0011 per session on Opus 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 2026-08-31.
Other skills, from other repositories
github-pr-creation
Creates GitHub Pull Requests with automated validation and task tracking. Use when user wants to create PR, open pull request, submit for review, or check if ready for PR. Analyzes commits, validates task completion, generates Conventional Commits title and description, suggests labels. NOTE - for merging existing…
babysit
PR babysitter: monitors CI status, auto-rebases when behind, auto-fixes CI where possible, delegates review comment handling to dlc:pr-check, and re-requests review after fixes. Designed for /loop usage with Remote Control.
stage
Use when committing work from the current session to stage ONLY hunks the session touched, not the entire file. Prevents accidentally staging unrelated uncommitted changes from other work.
gitplan
Plan and execute coherent Conventional Commit groupings for tangled working tree changes — multiple intertwined logical edits that need to be split into separate, reviewable commits.
review-staged
Top-level, user-facing workflow to review the staged Git diff for verified bugs and then prepare a clean Conventional Commit. Use when the user directly asks to review staged changes or prepare their commit. Do not use for a delegated leaf review or finding-validation task.
open-pr
Use this skill to submit the commits on the current branch as one pull request against a GitHub repository. Push the branch, open the PR with an imperative title and a body that names what changed and why, and post a follow-up ping only when CODEOWNERS resolves to one person. One branch per run, one pull request per…