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 skills/zerosixty/samurai/skillnpx skills add zerosixty/samurai --skill skillgit clone --depth 1 https://github.com/zerosixty/samuraiWhat 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 | $0.00069 | $0.01769 |
| Opus 5 | $0.00034 | $0.00885 |
| Sonnet 5 | $0.00014 | $0.00354 |
| Haiku 4.5 | $0.00007 | $0.00177 |
Grade A, and why
samurai 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 2d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Samurai Test Writing Rules
Good For: Go tests with a branching state-mutating tree (≥3 leaf paths, mutually exclusive siblings) using samurai.Run / samurai.RunWith.
Bad For: read-only chains, sequential accumulation, same-action-different-input variants, flat independent tests (see "When to Use" below for full criteria).
Rules
- One action + its assertions = one
Test(). Don't split assertions into child Tests — assert the result in the same callback that produced it. Child Tests are for new actions (setup, mutations, queries), not for checking fields of a parent's result - DECLARE variables with
varin the builder body, ASSIGN insideTest()callbacks - Each leaf path re-executes the builder from scratch — fresh variables, full isolation
w.Cleanup(fn): LIFO order, inner before outer, runs even on panic, thread-safes.Skip(): affects entire scope regardless of call order, no callbacks/cleanups executew.Context(): returns the scope'scontext.Context— use in factories for initialization needing a contextcontext.Context: first param inTestcallbacks, live during test, canceled after path completes- Parallel by default;
samurai.Sequential()forces order;go test -parallel Ncontrols concurrency - Assertion-agnostic: use
w.Testing()with any library (testify, is, stdlib)
Naming
Every s.Test("name", ...) — at every tree level — names a business action or outcome from the actor's POV. Not the function under test, the assertion, or the input shape.
Forbidden patterns:
- Identifier leak — CamelCase tokens matching Go identifiers in the Test body.
"UpdateQuote calls..."leaksUpdateQuote. - HTTP code leak — bare 3-digit codes (
\b[1-5]\d{2}\b) or4xx/5xxshorthand."returns 409"→"rejects duplicate id". - Assertion phrasing —
is / equals / returns / not nil / has length / contains."err is not nil"→"rejects invalid signature". - Structure phrasing —
empty / nil / with N / map / slice / struct."with empty slice"→"borrower has no past loans".
What ships with it
3 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.
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.
- 2d ago First seen · 84 lines · 69 tokens per session scan A b0caf6c6f028
samurai is a skill published in the GitHub repository zerosixty/samurai (11 stars, last pushed 3mo ago), licensed MIT. It adds 69 tokens to every session and 1,769 once invoked, about $0.0003 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 skills, from other repositories
dd-code-generation
Use pup CLI for immediate Datadog operations or generate code for integration into applications.
drt-analyze
Analyze DRT cluster health for a given time range. Reconstructs the operations timeline, checks CockroachDB metrics (availability, latency, storage, changefeeds, jobs, goroutines, admission control, LSM, KV prober) and logs for anomalies, correlates findings with disruptive operations to distinguish expected…
redux-to-swr
Migrate React components from Redux + Saga to SWR hooks. Use when converting data fetching from Redux store (reducers, sagas, selectors, connect HOC) to SWR-based hooks in CockroachDB DB Console or cluster-ui.
mma-investigator
Expert system for investigating MMA (Multi-Metric Allocator) behavior on CockroachDB clusters. Helps oncall engineers diagnose load imbalances, understand rebalancing decisions, and identify why MMA did or didn't act.
file-crdb-issue
Use when filing, creating, or reporting GitHub issues for CockroachDB. Use when asked to open a bug report, feature request, investigation issue, or performance inquiry. Also use when the user mentions wanting to track a problem, report a regression, or document unexpected behavior in CockroachDB.
engflow-artifacts
Use when downloading test logs, artifacts, or outputs.zip from EngFlow build invocations. Use when investigating CockroachDB CI test failures hosted on mesolite.cluster.engflow.com.