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/toverux/cantrips/commitnpx skills add toverux/cantrips --skill commitgit clone --depth 1 https://github.com/toverux/cantripsWhat 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 | $0.00028 | $0.01237 |
| Opus 5 | $0.00014 | $0.00619 |
| Sonnet 5 | $0.00006 | $0.00247 |
| Haiku 4.5 | $0.00003 | $0.00124 |
Grade A, and why
commit 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 yesterday.
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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Close the loop: harvest the session's learnings first, then create well-crafted git commits from the working tree — learning writes included, so the tree is clean when the loop ends.
Step 1: Scan for compound candidates
Scan the whole session for candidate learnings: root causes uncovered, gotchas hit, approaches that failed, conventions or preferences decided along the way — plus anything /review-gate flagged as compound material.
Judge each against compound's quality bar — would it change a future agent's behavior in a different session, and is it non-obvious and stable? Session-specific trivia dies here.
- Any candidate might clear the bar → invoke
/compoundwith the candidates; it routes each to a destination and gates every write on the user. Once the approved writes land, present their diff and wait for the user's verdict on the wording — approve, edit, or drop each document — since the destination gate cleared a one-line proposal, not the prose itself. Writes that survive ride into the commits below. - None → say so in one line and move on.
Step 2: Gather context
Gather context with each command as its own shell tool call (program + args only). Do not join with ;, &&, ||, pipes, $(...), or redirects — that syntax fails under Windows PowerShell. A non-zero exit is a normal state to interpret, not a failure to suppress.
git status— working-tree state. Clean tree → report there is nothing to commit and stop.git diff HEAD— the uncommitted changes.git branch --show-current— empty output means detached HEAD.git log --oneline -10— recent commit style.git rev-parse --abbrev-ref origin/HEAD— the default branch (strip theorigin/prefix). If unset, fall back tomain.
Treat this as a snapshot. Re-read branch and staged set immediately before committing if anything may have changed.
Step 3: Choose the branch
Detached HEAD → cut a feature branch before committing, and do not ask — even where the user asked only for a commit, since a commit reachable from nothing is lost as soon as HEAD moves. On the default branch, follow the repo's workflow instead: where the documented conventions or recent history show feature branches (or the user asked for one), branch off it before committing. Where the history shows commits landing directly on it (trunk-based), commit where you are. On any other branch, commit where you are unless the user asked for a new one.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 86 lines · 28 tokens per session scan A 3630ee227aae
commit is a skill published in the GitHub repository toverux/cantrips (2 stars, last pushed 2d ago), licensed MIT. It adds 28 tokens to every session and 1,237 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 2026-08-31.
Other skills, from other repositories
create-skill
Creates new skills from the user's repeating patterns. Interview-driven: discovers the task, analyzes failure modes, generates a production SKILL.md, installs it, tests it on a real target, and teaches the user how to use it. Use when a user wants to encode a repeating workflow; do NOT use for one-off tasks or…
map-learn
Capture reusable lessons after a completed MAP workflow. Use when a MAP run has finished and you want rules written to .claude/rules/learned/ from a workflow summary or handoff. Do NOT use during active implementation.
skills-management
Search, find, discover, install, remove, update, review, deduplicate, list, move, optimise, and iterate on skills for AI coding agents. Use when user asks "find a skill for X", "install skill", "remove skill", "update skills", "list skills", "deduplicate skills", "why are two skills shown", "choose the canonical…
map-understand
Interactive deep-understanding and quiz mode for MAP sessions. Use when the user wants to understand code, a diff, workflow result, debugging cause, or architecture and be checked with restatements or quizzes.
finance-invest-primer
투자를 처음 시작하는 사람에게 기본 원칙과 자산군, 초보 포트폴리오를 안내하는 스킬입니다. '투자 어떻게 시작해', '주식 초보', 'ETF가 뭐야', '분산투자 방법', '초보 포트폴리오 짜줘'처럼 말하면 됩니다. 분산·장기·리스크 관리 원칙, 자산군(주식/ETF/채권/부동산/예적금) 비교, 초보 포트폴리오, 투자 사기 회피를 한국 투자자 관점에서 설명합니다.
course-creator
Interactive, step-by-step course creation assistant. Takes any idea (or YouTube video) and produces a complete, launch-ready course package. USE THIS SKILL WHEN: User mentions creating a course, building a course, course creation User wants a course outline, curriculum, or course structure User says "turn this into a…