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 skills add triumsebas/unlimited-mcp --skill unlimited-mcp-selftestgit clone --depth 1 https://github.com/triumsebas/unlimited-mcpWrote 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/triumsebas/unlimited-mcp/unlimited-mcp-selftest)<a href="https://agentmods.dev/skills/triumsebas/unlimited-mcp/unlimited-mcp-selftest"><img src="https://agentmods.dev/badge/skills/triumsebas/unlimited-mcp/unlimited-mcp-selftest/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/triumsebas/unlimited-mcp/unlimited-mcp-selftest"><img src="https://agentmods.dev/badge/skills/triumsebas/unlimited-mcp/unlimited-mcp-selftest.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 | $0.00098 | $0.01602 |
| Opus 5 | $0.00049 | $0.00801 |
| Sonnet 5 | $0.00020 | $0.00320 |
| Haiku 4.5 | $0.00010 | $0.00160 |
Grade C, and why
unlimited-mcp-selftest scanned grade C with 1 finding 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 12d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf /tmp/umcp-selftest How it starts
The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: unlimited-mcp-selftest
Claude-Code wrapper that executes the project's regression battery and returns a single pass/fail report. The canonical test definitions live in AGENTS.md → "Regression suite"; this file only adds the execution flow.
Abort as soon as any tier fails — do not proceed to the next one.
Safety rule: never use the user's repos as test targets. All coding
agent tests (B2, B4, B6b) use throwaway git repos created in /tmp and
deleted immediately after.
Preparation (before Tier B)
Create two throwaway git repos — one local, one on the remote:
# local throwaway repo
mkdir -p /tmp/umcp-selftest
cd /tmp/umcp-selftest
git init
echo 'def hello():\n pass' > hello.py
git add hello.py
git commit -m "init"
# remote throwaway repo (via run_command on mcp_localhost)
run_command(["git", "init", "/tmp/umcp-selftest"], exec_host="mcp_localhost")
run_command(["git", "-C", "/tmp/umcp-selftest", "commit", "--allow-empty", "-m", "init"],
exec_host="mcp_localhost")
# write hello.py on remote via a small echo chain or sftp
Also: add_allowed_root("/tmp/umcp-selftest") so the agent can access it.
Flow
0. Tier 0 — Lint & types (headless, always)
Run from the unlimited-mcp repo root (not the throwaway repo):
uv run ruff check src tests
uv run ruff format --check src tests
uv run mypy
All green → continue. Any failure → stop, report output.
1. Tier A — Automated (always)
uv run pytest tests/integration/test_smoke.py -q
All green → continue. Any failure → stop, report failing test + assertion.
2. Tier B — Live end-to-end
B1 — Sysops remote
r = run_command(["echo", "b-remote"], exec_host="mcp_localhost")
r = await_job(r["job_id"])
Pass: status == "completed" and stdout contains b-remote.
B2 — Agent local + worktree (uses /tmp/umcp-selftest)
r = delegate_to_agent(
"opencode_flash",
prompt='Add a one-line docstring to the hello() function in hello.py. Only change that file.',
cwd="/tmp/umcp-selftest",
workspace="safe_dev",
timeout_seconds=180,
)
r = await_job(r["job_id"])
Pass: status == "completed", branch non-null, changed_files non-empty.
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.
- 12d ago First seen · 192 lines · 98 tokens per session scan C b0d6fe1aeae2
unlimited-mcp-selftest is a skill published in the GitHub repository triumsebas/unlimited-mcp (2 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 98 tokens to every session and 1,602 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
e2e
Selects and runs the appropriate AgentsMesh end-to-end suite for Web, Desktop, MCP, or iOS, including worktree-specific environment setup and browser-level verification. Use when a change needs E2E coverage, a user asks to execute or diagnose an E2E test, or a cross-service workflow must be verified against the real…
e2e-testing
Playwright E2E testing patterns, Page Object Model, configuration, CI/CD integration, artifact management, and flaky test strategies.
ai-regression-testing
Regression testing strategies for AI-assisted development. Sandbox-mode API testing without database dependencies, automated bug-check workflows, and patterns to catch AI blind spots where the same model writes and reviews code.
automated-web-interactive-bug-reporting
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
diagnose
Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.
tdd-workflow
Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.