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 mikestangdevs/craft-skills --skill bank-the-suitegit clone --depth 1 https://github.com/mikestangdevs/craft-skillsWrote 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/mikestangdevs/craft-skills/bank-the-suite)<a href="https://agentmods.dev/skills/mikestangdevs/craft-skills/bank-the-suite"><img src="https://agentmods.dev/badge/skills/mikestangdevs/craft-skills/bank-the-suite/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/mikestangdevs/craft-skills/bank-the-suite"><img src="https://agentmods.dev/badge/skills/mikestangdevs/craft-skills/bank-the-suite.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 | $0.00141 | $0.01196 |
| Opus 5 | $0.00071 | $0.00598 |
| Sonnet 5 | $0.00028 | $0.00239 |
| Haiku 4.5 | $0.00014 | $0.00120 |
Grade A, and why
bank-the-suite 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 11d 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bank the Suite
The failure mode this fixes
The agent changes a module, runs the three tests next to it, sees green, and moves on. Four tasks later the full suite runs for the first time and 23 tests are red — and now nobody knows which of the four changes broke them, the context that would have made the diagnosis cheap is gone, and the "quick fix" becomes an archaeology dig.
The inverse failure is just as common: the suite is run, 14 tests fail, and the report says "some pre-existing failures, probably unrelated." Probably. Unverified red is a debt that compounds — every subsequent change hides behind it.
This skill makes the full suite a bank: after each unit of work you deposit a verified green state, so the next unit starts from known-good and any new red has exactly one suspect.
When to Use This Skill
- A feature, fix, or refactor just landed and the next task is queued
- Only the tests adjacent to the change were run
- You're about to compact context, hand off, or end the session — the next session needs a trusted baseline
- The suite is slow and keeps getting skipped because "it takes 20 minutes"
- A report says failures are "pre-existing" or "unrelated" without naming them
Don't use when: the repo has no meaningful suite (then the gap is the finding — say so), or mid-exploration when nothing has landed yet. Don't block a one-character doc fix on a 20-minute suite — scale the gate to the blast radius.
Instructions
1. Establish the baseline once
Before the first change of the session, know what the suite looks like clean: total collected, passing, and any already-failing tests by name. If you skipped this and the suite is now red, check the failing tests against the main branch / pre-change state before assuming your change caused it — but check, don't assume.
2. Run the whole thing, not the neighborhood
Tests near the change verify the change; the rest of the suite verifies the blast radius — the callers, the shared fixtures, the serialization someone else depends on. The failures that hurt are always in code you didn't think you touched. If a full run is genuinely too slow for every iteration, run the affected subset per iteration and the full suite at the unit boundary — but the unit doesn't close until the full run is green.
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.
- 11d ago First seen · 76 lines · 141 tokens per session scan A 17405eef4ed6
bank-the-suite is a skill published in the GitHub repository mikestangdevs/craft-skills (4 stars, last pushed 3mo ago), licensed MIT. It adds 141 tokens to every session and 1,196 once invoked, about $0.0007 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 2026-08-31.
Other skills, from other repositories
craft-testing
Craftsman standard for automated testing: strategy, unit/integration/e2e selection, refactor-proof design, flaky tests, mocking boundaries, deterministic data, and merge-gate policy. Use WHENEVER work touches tests: writing/reviewing tests, strategy, "add tests", "why is this flaky", "what should I test", "tests pass…
integration-e2e-testing
Integration and E2E test design principles, ROI calculation, test skeleton specification, and review criteria. Use when designing integration tests, E2E tests, or reviewing test quality.
working-effectively-with-legacy-code
Apply Michael Feathers-inspired legacy-code rules when changing hard-to-test code safely with characterization tests, seams, sprout methods, or wrap methods.
refactoring-guru
Apply Refactoring.Guru-inspired rules when diagnosing code smells, choosing refactoring techniques, and stopping cleanup before uncontrolled redesign.
refactoring
Apply Martin Fowler-inspired refactoring rules when improving existing code structure while preserving observable behavior.
lab:autoresearch
Self-improving loop for plugin skills. Reads program.md, proposes one mutation per iteration, evaluates against deterministic scorer, keeps improvements via git, reverts failures. Targets weakest skill+dimension. Use with /loop for overnight runs.