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/mjmorales/claude-prove/commitnpx skills add mjmorales/claude-prove --skill commitgit clone --depth 1 https://github.com/mjmorales/claude-proveWhat 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 | $0.00037 | $0.00938 |
| Opus 5 | $0.00018 | $0.00469 |
| Sonnet 5 | $0.00007 | $0.00188 |
| Haiku 4.5 | $0.00004 | $0.00094 |
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 2d 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 — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Semantic Commit Assistant
Rules
- Never force push or amend existing commits. Instead, create new commits.
- Never hardcode scopes. Instead, derive from
.claude/.prove.jsonor directory structure. - Never ask about a scope the user has declined (recorded in a
declined_scope_<name>.mdmemory file). Instead, silently use directory-derived scope. - Delegate all user decisions through
AskUserQuestionperreferences/interaction-patterns.md.
1. Gather Context
Run in parallel:
- Read
.claude/.prove.json-- extract the"scopes"key if present git status,git diff,git diff --cached
If no changes exist, inform the user and stop.
2. Derive Scopes
With configured scopes (.claude/.prove.json has "scopes"):
Match each changed file to the scope whose path prefix is the longest match.
{ "scopes": { "api": "src/api/", "auth": "src/auth/", "db": "src/models/" } }
Without configured scopes: derive from the most meaningful path segment (e.g., src/api/handler.go -> api).
Built-in scopes (always available, override nothing):
| Scope | Matches |
|---|---|
docs |
README, LICENSE, top-level documentation |
repo |
.gitignore, CI/CD, repo infrastructure |
config |
.claude/.prove.json, project config files |
3. Detect Scope Gaps
Skip if .claude/.prove.json has no "scopes" key.
If changed files fall outside ALL configured scope prefixes AND do not match a built-in scope:
- Determine an appropriate scope name and path prefix for the unmatched files
- Check
~/.claude/projects/*/memory/fordeclined_scope_<name>.mdfor this project - Decline on record: use directory-derived scope silently, proceed to step 4
- Not declined: use
AskUserQuestion:
AskUserQuestion:
question: "Files under `<path>` don't match any configured scope. Add `<name>: <prefix>` to .claude/.prove.json?"
header: "New Scope"
options:
- label: "Add Scope"
description: "Register in .claude/.prove.json and use it for this commit"
- label: "Skip"
description: "Use directory-derived scope; never ask about this scope 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.
- 2d ago First seen · 110 lines · 37 tokens per session scan A a5f3235e3e91
commit is a skill published in the GitHub repository mjmorales/claude-prove (2 stars, last pushed 22d ago), licensed MIT. It adds 37 tokens to every session and 938 once invoked, about $0.0002 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
git-workflow-and-versioning
Structures git workflow practices. Use when making any code change. Use when committing, branching, resolving conflicts, opening or reviewing a pull request (PR), pushing to a remote, or when you need to organize work across multiple parallel streams. Use when cutting a release, choosing a semantic version bump…
prowler-commit
Creates professional git commits following conventional-commits format. Trigger: When creating commits, after completing code changes, when user asks to commit.
keploy-pr-workflow
Guide for creating PRs and issues on keploy repositories — PR format, customer-data hygiene, commit conventions, sign-off. Invoke when the user asks to open, update, or review a pull request or issue, when preparing a commit that will land in main, or whenever a change is about to leave the local machine.
git-commit
Generate well-formatted git commit messages following conventional commit standards.
git-workflow
This skill should be used when the user asks to "create git commit", "manage branches", "follow git workflow", "use Conventional Commits", "handle merge conflicts", or asks about git branching strategies, version control best practices, pull request workflows. Provides comprehensive Git workflow guidance for team…
commits
Use this skill whenever the user asks to commit changes, write or fix a commit message, amend or rename a commit, or do a workflow that includes committing in the IntelliJ repository. This is a thin repo-specific overlay: use IntelliJ commit format, write full commit messages by default, and keep requested suffixes…