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 agentmods add skills/truecourse-ai/truecourse/truecourse-fixnpx skills add truecourse-ai/truecourse --skill truecourse-fixgit clone --depth 1 https://github.com/truecourse-ai/truecourseWrote 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/truecourse-ai/truecourse/truecourse-fix)<a href="https://agentmods.dev/skills/truecourse-ai/truecourse/truecourse-fix"><img src="https://agentmods.dev/badge/skills/truecourse-ai/truecourse/truecourse-fix.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 | $0.00013 | $0.00750 |
| Opus 5 | $0.00006 | $0.00375 |
| Sonnet 5 | $0.00003 | $0.00150 |
| Haiku 4.5 | $0.00001 | $0.00075 |
Grade A, and why
truecourse-fix 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 6d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TrueCourse Fix
Apply fixes for TrueCourse violations that have fix suggestions.
Important
- Always invoke via
npx -y— without-y, npx will hang on the "Ok to proceed?" prompt whenever the user hasn't cached the latesttruecourseversion. - Default to the diff flow. Users usually want to fix the violations introduced by the changes they're currently iterating on — not the whole repo. Start by asking which set to fix.
- Only violations with a
Fix:block can be auto-fixed. Violations without one require human design decisions; mention them but don't attempt them.
Instructions
1. Pick the violation set
Ask: "Do you want to fix violations from the latest full analysis, or just the changes you're working on right now (diff)?"
-
Diff mode (recommended default):
npx -y truecourse list --diff— small set, usually fine to load in one call. -
Full mode: start with a paged view,
npx -y truecourse list(first 20). If the summary line shows a large total and the user hasn't narrowed scope, ask them to narrow before pulling more pages:- by severity — use
--severity <list>, e.g.npx -y truecourse list --severity critical,high. Valid values:critical,high,medium,low,info. - by a specific file / module / service they care about — no server-side filter yet, so page through and keep only matches.
Avoid
--allon large repos — it dumps every violation into context even though most won't be fixable. - by severity — use
If list --diff returns "no diff results yet" or "stale diff", suggest the user first run /truecourse-analyze in diff mode.
2. Identify fixable violations
From the output in hand, keep only violations that contain a Fix: block — those are the ones with actionable fix suggestions. If you loaded only one page in full mode, tell the user how many were fixable on this page and offer to continue to the next page (--offset 20, --offset 40, …) if they want more.
If none on the loaded page(s) are fixable, tell the user and stop (or offer to page further).
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.
- 6d ago First seen · 61 lines · 13 tokens per session scan A a28d5294f907
truecourse-fix is a skill published in the GitHub repository truecourse-ai/truecourse (527 stars, last pushed yesterday), licensed MIT. It adds 13 tokens to every session and 750 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
implement
Research, implement, test, document, and review a Fallow feature, fix, refactor, or repository improvement. Use when asked to build or change Fallow.
open-draft-pr
Prepare local changes for review with an intentional commit, push, and ready PR for fallow. Use when the user wants to publish work, open a PR, or turn local changes into a reviewable branch.
browser-smoke-review
Use browser automation to review docs pages, preview URLs, rendered output, or web-facing fallow surfaces. Use when the user wants a screenshot-based review, browser smoke test, docs site check, or preview deployment inspection.
panel-review-loop
Iteratively review and improve a Fallow user-facing surface across representative real-world projects until the panel has no blocking concerns.
review
Perform Fallow's comprehensive pre-merge review. Use after implementation or when asked to review a branch, pull request, or diff.
address-pr-comments
Triage and address GitHub PR review feedback for fallow, then implement the agreed fixes. Use when the user wants to inspect PR comments, requested changes, or unresolved review threads and act on them.