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 commands/fprochazka/claude-code-plugins/mr-opengit clone --depth 1 https://github.com/fprochazka/claude-code-pluginsWhat 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 | $0.00014 | $0.00688 |
| Opus 5 | $0.00007 | $0.00344 |
| Sonnet 5 | $0.00003 | $0.00138 |
| Haiku 4.5 | $0.00001 | $0.00069 |
Grade A, and why
mr-open 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 — 32 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Open MR
Open a draft merge/pull request for the current branch, then refine the title and description so they accurately describe the actual changes.
$ARGUMENTS
Process
- Load skills — invoke the
git:git-workflowskill for commit/MR shaping judgment. Also invoke the skill that covers the CLI for the remote host (GitHub, GitLab, …) so you don't guess at its syntax. - Open the draft MR — push the branch, then open the merge/pull request as a draft, targeting the repository's base branch and assigned to the user. Prefer whatever the project or the user's setup already provides for this — a dedicated script or alias that derives the title from the ticket, otherwise the host's CLI directly. If an MR/PR for this branch already exists, take its URL and proceed to step 3.
- Inspect what actually changed — review the diff against the base branch and the commit history. Don't trust the original ticket title/description verbatim; the implementation may have diverged.
- Write an accurate title — short, imperative, specific. Keep the repository's title convention, including any
<ticket-id>:prefix already in place. Update the rest to describe what the MR actually does. - Write a description for the reviewer — the goal is to set the reviewer up with the right optics for reading the diff, not to recap what changed (the diff already shows that). Cover, as applicable:
- Why — the motivation / problem being solved, the constraint or decision that shaped the approach. If the ticket already captures this well, a one-line pointer is enough.
- Gotchas — non-obvious tradeoffs, things that look wrong but aren't, assumptions that depend on external state, anything a reviewer might flag as a bug without context.
- Where to focus — call out the 1-3 most complex / load-bearing parts of the diff (with file:line pointers) that deserve the most careful review. Trivial parts (renames, formatting, mechanical refactors) can be explicitly de-prioritized.
- Keep it tight. No empty sections, no filler, no "already done in branch X" meta-commentary.
- Update the MR — write the new description to a file in the session scratchpad dir (e.g.
<scratchpad>/<ticket-or-branch>-mr-description.md), then set it through the host's CLI by reading that file into the argument rather than inlining the text. Update the title the same way if it changed. - Report — print the MR URL and a one-line summary of what you set.
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 · 32 lines · 14 tokens per session scan A fa2dd4b8a7ef
mr-open is a command published in the GitHub repository fprochazka/claude-code-plugins (11 stars, last pushed 4d ago), licensed MIT. It adds 14 tokens to every session and 688 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 commands, from other repositories
review-renovate
Review and merge renovate PRs with automerge configuration updates.
push-and-release
Git pull, resolve conflicts, push, fix hook errors, then release.
release
Perform a full release: generate release notes, bump version, commit, tag, and push.
release
Create a full stable release for the project. Optionally pass a title suffix (e.g., "Template Presets Fix") or extra notes.
SHIP_RELEASE
Ship the active release/X.Y.Z — build the installer, merge to main, tag, push, and publish the GitHub release. Run after /prerelease passes.
release
The pre-tag controller: version sync, changelog, clean tree, gate, and suite — every failure listed, the tag printed, never run.