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/monte-carlo-data/mc-agent-toolkit/mc-validategit clone --depth 1 https://github.com/monte-carlo-data/mc-agent-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/commands/monte-carlo-data/mc-agent-toolkit/mc-validate)<a href="https://agentmods.dev/commands/monte-carlo-data/mc-agent-toolkit/mc-validate"><img src="https://agentmods.dev/badge/commands/monte-carlo-data/mc-agent-toolkit/mc-validate.svg" alt="Measured on agentmods" 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 | $0.00009 | $0.00586 |
| Opus 5 | $0.00005 | $0.00293 |
| Sonnet 5 | $0.00002 | $0.00117 |
| Haiku 4.5 | $0.00001 | $0.00059 |
Grade A, and why
mc-validate 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 4d 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.
What it actually says
Parse $ARGUMENTS to decide which mode to run. The Monte Carlo Prevent skill defines four supported invocations:
| Invocation | Mode | Workflows |
|---|---|---|
| (empty) | default generate | W3 only |
generate |
explicit generate | W3 only |
run |
full validate | W4.1 (Build) → W4.2 (Execute) |
run --skip-build |
execute-only | W4.2 only |
run also accepts --dev-db <NAME> to bypass the dev-database prompt in W4.2. Flags can combine: run --skip-build --dev-db <NAME>.
run does not auto-generate. If no validation/<table>_<ts>.sql exists for the current session's changed models, abort and tell the engineer:
"No validation queries found for <table_name>. Run
/mc-validate(or/mc-validate generate) first to generate them, then re-run/mc-validate run."
Decision rules:
-
If
$ARGUMENTSis empty or starts withgenerate: generate-only mode. Run Workflow 3 from the Monte Carlo Prevent skill. Save queries tovalidation/<table_name>_<timestamp>.sql. At the end, always offer the four next-step options described in W3 step 6 ofreferences/workflows.md(say continue to auto-run/mc-validate run, run it yourself, run with--skip-build, or substitute<YOUR_DEV_DATABASE>and run manually). -
If
$ARGUMENTSstarts withrun: run mode.- Verify a
validation/<table>_<ts>.sqlexists for the current session's changed models. If none exist, abort with the message above. Do not run W3. - If
--skip-buildis not present: run Workflow 4.1 (Build), then Workflow 4.2 (Execute). - If
--skip-buildis present: skip W4.1 and run Workflow 4.2 directly. Assume the engineer built manually. - If
--dev-db <NAME>is present: pass<NAME>into W4.2 and skip its dev-database confirmation prompt.
- Verify a
-
Any other token after
runthat isn't--skip-buildor--dev-db <NAME>is an error — surface it to the engineer and stop.
Workflow definitions live in the Monte Carlo Prevent skill's references/workflows.md. Read that file before executing either mode.
Arguments: $ARGUMENTS
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.
- 4d ago First seen · 35 lines · 9 tokens per session scan A 262cdeae7eac
mc-validate is a command published in the GitHub repository monte-carlo-data/mc-agent-toolkit (91 stars, last pushed 10d ago), licensed Apache-2.0. It adds 9 tokens to every session and 586 once invoked, about $0.0000 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
cleanup
Detect and remove orphaned code, unused components, dead routes, and stale database artifacts.
mcp-guide
Reference for the MCP toolset (Context7, Firecrawl, Playwright, Supabase, Sequential Thinking) and when to reach for each.
reflect
Reflect on an outcome to update memory confidence and create relationships in the central Gnosys database.
rls-plan
Supabase RLS + access-control audit + remediation plan — no SQL until approved.
security-plan
OWASP Top 10 + Supabase security audit + hardening plan — no fixes until approved.
pn-audit-data
Design or review database schema — normalization, indexing strategy, migration plan, and data model decisions. Surgical command for database design. Use standalone or as part of pn-backend-audit.