write-tests

A testing guide for Skillprobe, a tool that checks how language-model skills behave and when they activate. It creates YAML files describing these tests.

In plain words
What is it for?
It helps test a skill's responses and check that it activates for relevant prompts while staying inactive for unrelated ones.
Why use it?
It removes the need to design Skillprobe test files from scratch or guess whether a skill responds to the right requests.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/anyesh/skillprobe/write-tests
Any agent
npx skills add Anyesh/skillprobe --skill write-tests
Clone the repo
git clone --depth 1 https://github.com/Anyesh/skillprobe

Made for: Claude Code, Codex.

Per session 41 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,448 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00041 $0.02448
Opus 5 $0.00020 $0.01224
Sonnet 5 $0.00008 $0.00490
Haiku 4.5 $0.00004 $0.00245

Measured 2d ago against content hash 6a952c166072, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

write-tests 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.

.agents/skills/write-tests/SKILL.md · 249 lines

How it starts

The opening of the file, as written. The whole thing — 249 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Writing skillprobe Tests

Generate YAML test files for skillprobe. Always read the skill content first before writing tests.

When asked to write tests for a skill, generate BOTH files unless the user asks for only one:

  1. A behavioral test file (for skillprobe run)
  2. An activation test file (for skillprobe activation)

Behavioral Tests (for skillprobe run)

Test what the model DOES when the skill is loaded.

harness: claude-code    # or cursor
model: claude-haiku-4-5-20251001
timeout: 120
skill: ./path/to/skill  # directory with SKILL.md, or a .md file

scenarios:
  - name: "descriptive name"
    steps:
      - prompt: "what the user would type"
        assert:
          - type: contains
            value: "expected text"

Activation Tests (for skillprobe activation)

Test whether the skill LOADS for relevant prompts and stays quiet for irrelevant ones. This is a separate YAML format from behavioral tests.

harness: claude-code
model: claude-haiku-4-5-20251001
timeout: 60
skill: ./path/to/skill

activation:
  skill_name: my-skill
  should_activate:
    - "prompt that should trigger this skill"
    - "another relevant prompt"
  should_not_activate:
    - "completely unrelated prompt"
    - "another unrelated prompt"

The activation key replaces scenarios. These are two different file formats loaded by different commands (skillprobe run vs skillprobe activation). Never put both keys in the same file. skillprobe will either ignore one silently or fail to parse the file.

Bad:

skill: ./skills/commit
activation:
  skill_name: commit
  should_activate:
    - "commit my changes"
scenarios:
  - name: "commit happens"
    steps:
      - prompt: "commit"
        assert:
          - type: contains
            value: "commit"

Pick one format per file.

Testing skill combinations

Some tests need to load more than one skill at a time, either because you are probing interactions between skills or because the scenario legitimately needs both. Use a skills: list at the suite level instead of the singular skill::

Read the full file on GitHub · 249 lines

Changes

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.

  1. 2d ago First seen · 249 lines · 41 tokens per session scan A 6a952c166072

Subscribe to this mod's changes

write-tests is a skill published in the GitHub repository Anyesh/skillprobe (5 stars, last pushed 4mo ago), licensed MIT. It adds 41 tokens to every session and 2,448 once invoked, about $0.0002 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.

Related

Other skills, from other repositories

crossing-the-chasm

Navigate the technology adoption lifecycle from early adopters to mainstream market. Use when the user mentions "crossing the chasm", "beachhead segment", "whole product", "early adopters vs mainstream", "tech go-to-market", "bowling pin strategy", "technology adoption lifecycle", "pragmatist buyers", "growth stalled…

wondelai/skills · 139 tokens

design-everyday-things

Apply foundational design principles: affordances, signifiers, constraints, feedback, and conceptual models. Use when the user mentions "why is this confusing", "affordance", "error prevention", "discoverability", "human-centered design", "mental model", "mapping", "seven stages of action", "users keep making…

wondelai/skills · 132 tokens

hooked-ux

Design habit-forming product loops using the Hook Model (Trigger, Action, Variable Reward, Investment). Use when the user mentions "users arent coming back", "habit formation", "engagement loops", "habit zone", or "the manipulation matrix". Also trigger when designing notification or re-engagement strategies, building…

wondelai/skills · 109 tokens

architecture-optimization

Guided journey from a working codebase grown slow and tangled to one measurably fast, cleanly bounded, and readable. Orchestrates eight skills phase by phase - working-with-legacy-code, clean-architecture, software-design-philosophy, refactoring-patterns, system-design, ddia-systems, release-it, pragmatic-programmer …

wondelai/skills · 226 tokens

create-app

Guided journey from a raw app idea to a validated, cleanly architected first version that ships on a sustainable cadence. Orchestrates ten skills phase by phase - lean-startup, design-sprint, clean-architecture, domain-driven-design, clean-code, pragmatic-programmer, system-design, ios-hig-design, 37signals-way…

wondelai/skills · 217 tokens

create-business

Guided journey from raw idea to a validated, positioned, priced business with a chosen beachhead. Orchestrates ten skills phase by phase - jobs-to-be-done, mom-test, design-sprint, lean-startup, good-strategy-bad-strategy, blue-ocean-strategy, obviously-awesome, hundred-million-offers, monetizing-innovation…

wondelai/skills · 211 tokens