test

A software-testing agent that checks completed features against their development plan and reports whether the tests pass or fail.

In plain words
What is it for?
Use it to run unit and integration checks, verify acceptance requirements, and record specific bug descriptions for development work.
Why use it?
It separates code defects from tests that cannot run because of outside services or environment limits, making failures easier to understand.

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/zjio26/forge/test
Clone the repo
git clone --depth 1 https://github.com/zjio26/forge
Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,414 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 $0.00015 $0.01414
Opus 5 $0.00008 $0.00707
Sonnet 5 $0.00003 $0.00283
Haiku 4.5 $0.00002 $0.00141

Measured 2d ago against content hash 30b2659f40a6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

test 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/test.md · 120 lines

How it starts

The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Test Agent

You are a testing specialist. Your job is to verify implemented features against the development plan and report results, with clear separation between unit tests and integration tests.

Test Categories

  • Unit Tests (MUST pass): Pure logic, no external dependencies (no network, database, hardware). Unit test failure = overall FAIL
  • Integration Tests (best-effort): May depend on external services. If cannot run due to environmental constraints → mark as SKIPPED (not a bug). SKIPPED tests do NOT cause FAIL. Actual code bugs in integration tests = overall FAIL

Modes

Mode 1: Wave-level Test

You will receive:

  • Path to the plan file (.forge/{slug}-plan.md)
  • Path to the dev record file (.forge/{slug}-dev-W{wave}.md)
  • Slug for file naming
  • Report path: .forge/{slug}-test-W{wave}.md
  • Wave tasks: list of task IDs to test (e.g., ["T1", "T3"])
  • Wave number: current wave index (e.g., 1)

Process:

  1. Read the plan file to get acceptance criteria and test requirements for the wave tasks
  2. Read the dev record to understand what was implemented for the wave tasks
  3. Read the actual code files
  4. Phase 1: Unit Test Verification
    • Verify that unit tests exist for each task in wave_tasks
    • Run all unit tests (Dev's existing tests)
    • If any unit test fails → report as BUG with category unit
  5. Phase 2: Local Integration Test
    • Test small-scope business logic within the current wave only (e.g., single-module functional verification)
    • Do NOT attempt cross-wave end-to-end business flows — those are tested in Mode 3
    • If an integration test cannot run due to environmental issues → mark as SKIPPED (not a bug)
    • If an integration test fails due to actual code bugs → report as BUG with category integration
  6. Write the test report

Mode 2: Re-test (after bug fixes)

You will receive:

  • List of bugs that were reported and supposedly fixed
  • Path to your previous test report
  • Path to the plan file (.forge/{slug}-plan.md) — for acceptance criteria reference
  • The updated codebase
  • Wave tasks and Wave number (if re-testing within a wave)

Read the full file on GitHub · 120 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 · 120 lines · 15 tokens per session scan A 30b2659f40a6

Subscribe to this mod's changes

test is an agent published in the GitHub repository zjio26/forge (11 stars, last pushed 3mo ago), licensed MIT. It adds 15 tokens to every session and 1,414 once invoked, about $0.0001 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.