Borrowing it
Nothing to install: this file belongs to vanssata/jetbrains-claude-subscription. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/vanssata/jetbrains-claude-subscription/master/.claude/skills/bump-acp/SKILL.mdgit clone --depth 1 https://github.com/vanssata/jetbrains-claude-subscriptionWrote 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/vanssata/jetbrains-claude-subscription/bump-acp)<a href="https://agentmods.dev/skills/vanssata/jetbrains-claude-subscription/bump-acp"><img src="https://agentmods.dev/badge/skills/vanssata/jetbrains-claude-subscription/bump-acp/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/vanssata/jetbrains-claude-subscription/bump-acp"><img src="https://agentmods.dev/badge/skills/vanssata/jetbrains-claude-subscription/bump-acp.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00063 | $0.00636 |
| Opus 5.5 | $0.00025 | $0.00254 |
| Sonnet 5.5 | $0.00013 | $0.00127 |
| Haiku 4.5 | $0.00006 | $0.00064 |
Grade A, and why
bump-acp 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 today.
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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bump the pinned ACP package
The version lives in four places that must not drift. Change all of them in one commit.
1. Pick the version
Use the version given as the argument. Without one, take latest:
npm view @agentclientprotocol/claude-agent-acp version engines time.modified
Stop and ask before going on if engines.node is no longer >=22: MIN_NODE_MAJOR in
NodeRuntimeResolver.kt and the README mirror that floor. A version published in the last
day or two has had no users yet; say so and prefer one that has.
2. Edit the four places
Replace the old @agentclientprotocol/claude-agent-acp@<old> with the new version in:
src/main/kotlin/dev/vanssa/claudeacp/ClaudeAcpSettings.kt—DEFAULT_PACKAGE_SPECtest/handshake.sh— thePACKAGE=default (test/models.shreads it from there)README.md— theacp.jsonexample under "What it writes"README.md— thepackageSpecrow of the settings table
Then confirm nothing was missed:
grep -rn "claude-agent-acp@" README.md src test
One hit is meant to stay behind unless you know better: the command under "The problem it
solves" in README.md shows what JetBrains bundles, not what this plugin pins. Change it
only when the version the IDE ships has been looked up.
3. Check
sh test/handshake.sh
sh test/models.sh
handshake.sh must print PASS: claude-ai-login offered. models.sh must print PASS, or
SKIP without a Claude login — a SKIP leaves the model dropdown unchecked, so say so.
In a remote shell (SSH_CONNECTION, NO_BROWSER or CLAUDE_CODE_REMOTE set) the package
offers claude-login instead and handshake.sh fails for that reason alone. Run it with
those variables unset (env -u CLAUDE_CODE_REMOTE sh test/handshake.sh).
Then build and run the unit tests as CLAUDE.md describes.
4. Commit
feat: bump claude-agent-acp to <version>, with what the new version brings in the body.
A bump changes the default for every user, so it needs a release to reach them — see the
release gotcha in CLAUDE.md.
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.
- today First seen · 65 lines · 63 tokens per session scan A cb090264846d
bump-acp is a skill published in the GitHub repository vanssata/jetbrains-claude-subscription (18 stars, last pushed today), licensed MIT. It adds 63 tokens to every session and 636 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-11-11.
bump-acp compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| bump-acpvanssata/jetbrains-claude-subscription | Skill | 18 | today | MIT | A | 63 | Claude Code |
| swiftui-devOrkas-AI/Orkas | Skill | 9.2k | yesterday | MIT | A | 3 | Claude Code, Codex |
| kotlin-multiplatformtravisjneuman/.claude | Skill | 108 | today | MIT | A | 54 | Claude Code |
| ios-developmenttravisjneuman/.claude | Skill | 108 | today | MIT | A | 43 | Claude Code |
| macos-nativetravisjneuman/.claude | Skill | 108 | today | MIT | A | 40 | Claude Code |
| agent-kotlin-experttravisjneuman/.claude | Skill | 108 | today | MIT | A | 85 | Claude Code |
| android-developmenttravisjneuman/.claude | Skill | 108 | today | MIT | A | 33 | Claude Code |