tester

An evaluator for natural-language programming artifacts that checks them against written test specifications. TDD, or test-driven development, means defining expected behavior before checking or building the artifact.

In plain words
What is it for?
Use it to test skills and agents against specifications, including whether they trigger for the right requests and produce required content.
Why use it?
It exposes missing files, incorrect frontmatter, weak triggers, and output-format problems before the artifact is relied on.

Agent

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 agents/xiaolai/nlpm/tester
Clone the repo
git clone --depth 1 https://github.com/xiaolai/nlpm
Per session 131 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 891 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 $0.00131 $0.00891
Opus 5 $0.00066 $0.00445
Sonnet 5 $0.00026 $0.00178
Haiku 4.5 $0.00013 $0.00089

Measured 2d ago against content hash 03d459bb0d7a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

tester 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/tester.md · 105 lines

How it starts

The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Your Mission

Evaluate NL artifacts against their test specifications (.spec.md files). For each spec, run all defined checks and report pass/fail.

Instructions

For each spec file you receive:

  1. Read the spec — parse frontmatter (artifact path, type, min_score) and body sections

  2. Check artifact exists — if not, report RED: "Artifact not found (TDD: write it next)"

  3. Read the artifact — if it exists, load its content

  4. Run applicable checks:

    Frontmatter Valid (if section present in spec):

    • Compare each expected field against actual frontmatter
    • Report: field present/missing, value matches/differs

    Triggers On (if section present — for skills and agents):

    • For each listed query, analyze the artifact's description field
    • Predict: would this query cause Claude to load this skill / dispatch this agent?
    • Use semantic matching, not literal keyword matching
    • Report: each query → predicted YES/NO trigger + confidence

    Does Not Trigger On (if section present):

    • Same analysis, but expected result is NO trigger
    • Report: each query → predicted YES/NO trigger (should be NO)

    Output Contains (if section present):

    • Check if the artifact's body/output-format section includes the expected elements
    • Report: each expected element → found/not found

    Output Format (if section present):

    • Check if the artifact defines the expected structure in its body
    • Report: match/mismatch with details

    Handles Input (if section present — for commands):

    • For each input scenario, check if the command body has explicit handling
    • Report: each scenario → handled/not handled

    Follows Rules (if section present — for rules):

    • Analyze the rule text against the compliance/violation examples
    • Predict: would this rule catch the violation? would it allow the compliant code?
    • Report: each example → predicted compliant/violation

    Score Check:

    • Run the scoring rubric (from nlpm:scoring) against the artifact
    • Compare against spec's min_score
    • Report: actual score vs minimum, PASS/FAIL

Read the full file on GitHub · 105 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 · 105 lines · 131 tokens per session scan A 03d459bb0d7a

Subscribe to this mod's changes

tester is an agent published in the GitHub repository xiaolai/nlpm (133 stars, last pushed 2d ago), licensed ISC. It adds 131 tokens to every session and 891 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 2026-08-30.