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 atretyak1985/swarmery --skill guardrailsgit clone --depth 1 https://github.com/atretyak1985/swarmeryWrote 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/atretyak1985/swarmery/guardrails)<a href="https://agentmods.dev/skills/atretyak1985/swarmery/guardrails"><img src="https://agentmods.dev/badge/skills/atretyak1985/swarmery/guardrails/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/atretyak1985/swarmery/guardrails"><img src="https://agentmods.dev/badge/skills/atretyak1985/swarmery/guardrails.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 | $0.00057 | $0.00712 |
| Opus 5.5 | $0.00023 | $0.00285 |
| Sonnet 5.5 | $0.00011 | $0.00142 |
| Haiku 4.5 | $0.00006 | $0.00071 |
Grade A, and why
guardrails 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
A risky action gets a structured go/no-go before it runs, derived from an
Impact × Reversibility matrix — not ad-hoc vibes. The output contract is
binary: APPROVED (possibly with conditions) or REJECTED (with the
reason) — never ambiguous.
The check
- Short-circuit: read-only actions (viewing files, running tests) are APPROVED immediately — no ceremony.
- Risk level = Impact (Low / High / Critical) × Reversible (Yes / No):
Low×Yes → Low; High×Yes → Medium; Low/High×No → High; Critical×anything →
Critical. The matrix and per-action guidance:
resources/risk-checks.md. - Deterministic checks for the affected stack (typecheck/lint/tests for
code, dry-run/lint for config,
--dry-runfor migrations where the tooling offers one). A failing deterministic check is an automatic REJECTED. - Rollback plan with actionable commands ("
DROP TABLE IF EXISTS x;", "git revert <sha>") — "revert the change" is not a plan. - Critical never auto-approves. Anything Critical (prod data loss potential, credential changes, irreversible deletes) escalates to the user regardless of checks.
Verdict line: GUARDRAIL: {APPROVED|REJECTED} | Risk: {level} | Checks: {pass/fail/skip} | Rollback: {one line}.
How to use
What it does
Derives a risk level for a proposed action from an Impact × Reversibility matrix, runs the stack's deterministic checks, and emits an unambiguous APPROVED/REJECTED call with a concrete rollback plan — escalating anything Critical to the user instead of auto-approving.
When to use it
Before migrations, bulk deletes or renames, deploy/infra config changes, force operations, or anything an orchestrator flags as hard to reverse — especially inside autonomous runs where no human watches each step.
How to invoke
@core:tech-lead loads it via skills: and applies it before risky delegations; load it manually (/guardrails context or read resources/risk-checks.md) when about to run a risky command yourself.
What ships with it
1 file 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.
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 · 56 lines · 57 tokens per session scan A 2f6981c2c7eb
guardrails is a skill published in the GitHub repository atretyak1985/swarmery (7 stars, last pushed today), licensed Apache-2.0. It adds 57 tokens to every session and 712 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
dependency-versions
MUST consult this skill before answering whenever the user's task involves external versioned dependencies — even if you think you can handle it directly. This applies to: checking if packages/tools are up to date, upgrading npm/pip/cargo/go dependencies, planning or writing CI/CD workflows (GitHub Actions, CircleCI…
finish
Ship phase — merge, PR, keep, or discard with pre-merge verification.
gws-quick-ref
Quick reference for gws CLI syntax — core patterns, key flags, pagination, field masks, and output formats.
excalidraw
Create or extend editable Excalidraw diagrams from descriptions or repository evidence using a local skeleton-to-scene converter.
style-ops
Maintain, validate or report on the Fakoli principles ledger, checking evidence pointers and keeping its Markdown projection synchronized.
gws-events-subscribe
Google Workspace Events: Subscribe to Workspace events and stream them as NDJSON.