Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add aiocean/claude-plugins/plugin install aio-claude-toolkitWrote 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/aiocean/claude-plugins/aio-patch-setup)<a href="https://agentmods.dev/skills/aiocean/claude-plugins/aio-patch-setup"><img src="https://agentmods.dev/badge/skills/aiocean/claude-plugins/aio-patch-setup/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/aiocean/claude-plugins/aio-patch-setup"><img src="https://agentmods.dev/badge/skills/aiocean/claude-plugins/aio-patch-setup.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 | $0.00079 | $0.00992 |
| Opus 5.5 | $0.00032 | $0.00397 |
| Sonnet 5.5 | $0.00016 | $0.00198 |
| Haiku 4.5 | $0.00008 | $0.00099 |
Grade A, and why
aio-patch-setup 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 1mo 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
aio-patch-setup — scaffold a new Claude-binary-patching project
Goal
Drop a complete, runnable patching pipeline into the target directory so the user can immediately:
- Edit
tools/pipeline/patches.json(start from empty schema or copy frompatches.json.example) - Run
/aio-claude-toolkit:aio-patch-extractto extract cli.js + natives from their installed claude - Run
/aio-claude-toolkit:aio-patch-compileto apply patches + recompile a custom claude binary - Run
/aio-claude-toolkit:aio-patch-runto exec the patched binary
What gets copied
| Path | Source | User-owned? |
|---|---|---|
tools/extract.sh |
templates/tools/extract.sh (synced from dirty-claude + ARCH patch) | No — generic |
tools/build.sh |
templates/tools/build.sh | No — generic |
tools/run.sh |
templates/tools/run.sh | No — generic |
tools/pipeline/*.py |
dirty-claude pipeline (extract_cli, patch_cli, resolve_symbols, inline_sources, extract_native_modules) | No — generic |
tools/pipeline/patches.json |
EMPTY schema template | YES — user content |
tools/pipeline/patches.json.example |
dirty-claude pinned reference | No — reference |
tools/pipeline/sources/*.example |
dirty-claude reference patch body | No — reference |
tools/cli-nav/*.cjs |
AST navigator (acorn-based) | No — generic |
docs/*.md |
curated generic guides (anchor strategy, drift recovery, etc.) | No — generic |
.gitignore |
sensible defaults (dist/, *.backup, pycache/) | No — generic |
CLAUDE.md.example |
template for your project's CLAUDE.md | No — template |
README.md.example |
template for your project's README | No — template |
.aio-patch-setup (hidden) |
sync metadata (VERSION) | No — generated |
NOT copied (opt-in via sibling skill): control/ — call /aio-claude-toolkit:aio-patch-control scaffold to install the reference HTTP control-channel sample.
Usage
/aio-claude-toolkit:aio-patch-setup [target_dir]
target_dirdefaults to CWD.FORCE=1env var: overwrite generic scripts in an existing scaffold. Does NOT overwritetools/pipeline/patches.jsonortools/pipeline/sources/*.js(your patch content).
What ships with it
46 files 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.
- scripts/arch-env-override.patch 3.4 KB
- scripts/check-deps.sh 1.3 KB runs code
- scripts/setup.sh 4.4 KB runs code
- scripts/sync-from-dirty-claude.sh 6.5 KB runs code
- scripts/walk-up.sh 1001 B runs code
- templates/.gitignore 207 B
- templates/CLAUDE.md.template 1.3 KB
- templates/docs/architecture.md 4.9 KB
- templates/docs/caveats.md 11 KB
- templates/docs/control-channel.md 20 KB
- templates/docs/glossary.md 5.8 KB
- templates/docs/native-modules.md 4.8 KB
- templates/docs/patches.md 7.6 KB
- templates/docs/pipeline.md 5.1 KB
- templates/docs/README.md 5.0 KB
- templates/docs/repatching-playbook.md 16 KB
- templates/docs/symbol-resolver.md 6.3 KB
- templates/docs/versioning.md 5.4 KB
- templates/README.md.template 436 B
- templates/tools/build.sh 3.2 KB runs code
- templates/tools/cli-nav/build-explorer.cjs 31 KB runs code
- templates/tools/cli-nav/find-anchors.cjs 3.5 KB runs code
- templates/tools/cli-nav/lib-resolve.cjs 3.4 KB runs code
- templates/tools/cli-nav/lib.cjs 5.5 KB runs code
- templates/tools/cli-nav/navigate.cjs 2.7 KB runs code
- templates/tools/cli-nav/package-lock.json 2.9 KB
- templates/tools/cli-nav/package.json 360 B
- templates/tools/cli-nav/verify-explorer.cjs 2.6 KB runs code
- templates/tools/extract.sh 3.7 KB runs code
- templates/tools/pipeline/extract_cli.py 5.7 KB runs code
- templates/tools/pipeline/extract_native_modules.py 5.7 KB runs code
- templates/tools/pipeline/inline_sources.py 2.2 KB runs code
- templates/tools/pipeline/patch_cli.py 9.9 KB runs code
- templates/tools/pipeline/patches.json 1.0 KB
- templates/tools/pipeline/patches.json.example 26 KB
- templates/tools/pipeline/README.md 1.4 KB
- templates/tools/pipeline/resolve_symbols.py 7.3 KB runs code
- templates/tools/pipeline/sources/dirty_control_channel.js.example 21 KB
- templates/tools/pipeline/sources/README.md 732 B
- templates/tools/pipeline/test_resolve_symbols.py 8.1 KB runs code
- templates/tools/run.sh 2.7 KB runs code
- templates/VERSION 224 B
- tests/smoke-e2e.sh 2.4 KB runs code
- tests/test-check-deps.sh 680 B runs code
- tests/test-walk-up-from-subdir.sh 1.2 KB runs code
- tests/test-walk-up.sh 2.0 KB runs code
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.
- 1mo ago First seen · 68 lines · 79 tokens per session scan A e6dff60badef
aio-patch-setup is a skill published in the GitHub repository aiocean/claude-plugins (4 stars, last pushed 1mo ago), licensed MIT. It adds 79 tokens to every session and 992 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 2026-08-31.
Other skills, from other repositories
eval-hooks
Audit Claude Code and Codex hooks for validity, safety, and correctness across settings.json, hooks.json, config.toml, plugin, and skill or agent frontmatter hooks. Resolves each command, checks event names, handler types, matchers, exit-code and JSON decision strategy, and timeouts against each host's documented…
check-cache-bugs
Audit Claude Code setup for cache bugs (CC#40524): sentinel, --resume/--continue, attribution header + ArkNill B3/B4/B5.
sandbox-unblock
Diagnostic protocol to run before reporting a sandbox blocker or asking for a configuration change. Eleven checks that eliminate false positives, then a report template the person holding the settings can act on. On one measured day, six of eight reported blockers turned out to be false, all from the same handful of…
investigate
Systematic root-cause debugging: find the cause before writing any fix.
qa
Systematic QA testing of a web application: diff-aware, tiered, with fix-and-verify loop.
optimize
Analyze and suggest performance improvements for code, queries, or systems.