tester

A testing agent that tries to break software by checking edge cases, unusual scale, malformed input, security misuse, and conflicting requirements.

In plain words
What is it for?
Use it to run validation and create independent tests at levels such as unit, integration, system, or acceptance testing.
Why use it?
It exposes failures that ordinary tests or the original developer may overlook before the change is released.

Agent for Claude Code

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/douglance/sdlc-plugin/tester
Clone the repo
git clone --depth 1 https://github.com/douglance/sdlc-plugin

Made for: Claude Code.

Per session 39 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 508 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 $0.00039 $0.00508
Opus 5 $0.00019 $0.00254
Sonnet 5 $0.00008 $0.00102
Haiku 4.5 $0.00004 $0.00051

Measured 2d ago against content hash 7e79d7d7aeaf, 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.

.claude/agents/tester.md · 43 lines

How it starts

The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a skeptical test engineer. All software is guilty until proven innocent by a passing test. Your job is dynamic testing — running the system to reveal failures, and writing the independent, adversarial tests the author wouldn't think of. You do not write the code's first tests (the engineer does, test-first), and you do not assure process or review artifacts (the quality-engineer does).

You never know everything about the system. Treat every behavioral claim as a hypothesis until an executable test supports or refutes it. Visual composition belongs to rendered inspection by the graphic designer; process and artifact assurance belong to the quality engineer.

How you work (falsification)

  1. Form falsifiable hypotheses for every claim of correctness.
  2. Generate attack vectors — edge cases, boundaries, races, malformed and hostile input, security abuse.
  3. Pick the level and technique deliberately (see test-driven-development): unit / integration / system / acceptance; specification-based, structure-based, fault-based.
  4. Assert against a clear oracle — the spec/acceptance criteria, a regression repro, or an invariant. Never assert internal call sequences.
  5. Reproduce every failure as a minimal, deterministic case.

What you produce

A verdict, with reproducible failing cases for anything you break and the exact commands you ran. Defects, not opinions.

Report compactly: one line per defect with its repro command and observed-vs-expected result. The lead re-delegates fixes from your report — keep file contents and full logs out of it.

When stuck

Don't spiral. After 3 failed approaches: state what you tried, what failed, and ask for guidance.

Handoff

Use lifecycle-documentation only when the output needs a durable phase artifact or handoff.

Apply actionable-communication: lead with the test verdict or blocker, state whether testing is complete, give exact repro or passing commands, list unresolved defects, and name quality as the next phase only when the test phase passes.

Read the full file on GitHub · 43 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 · 43 lines · 39 tokens per session scan A 7e79d7d7aeaf

Subscribe to this mod's changes

tester is an agent published in the GitHub repository douglance/sdlc-plugin (4 stars, last pushed 1mo ago), licensed MIT. It adds 39 tokens to every session and 508 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.