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 agentmods add commands/thierryn/fire-flow/fire-testgit clone --depth 1 https://github.com/ThierryN/fire-flowWrote 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/commands/thierryn/fire-flow/fire-test)<a href="https://agentmods.dev/commands/thierryn/fire-flow/fire-test"><img src="https://agentmods.dev/badge/commands/thierryn/fire-flow/fire-test.svg" alt="Measured on agentmods" 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 | $0.00017 | $0.04306 |
| Opus 5 | $0.00009 | $0.02153 |
| Sonnet 5 | $0.00003 | $0.00861 |
| Haiku 4.5 | $0.00002 | $0.00431 |
Grade C, and why
fire-test scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf "$TEST_DIR" How it starts
The opening of the file, as written. The whole thing — 521 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/fire-test - Dominion Flow Integration Tests
Purpose
Run the Dominion Flow plugin test suite to verify all commands and integrations work correctly. Provides comprehensive validation of the entire plugin ecosystem before deployment or after updates.
Usage
/fire-test # Run all tests (E2E + Integration)
/fire-test --e2e # Run E2E tests only (8 tests)
/fire-test --integration # Run integration tests only (3 tests)
/fire-test --command [name] # Test specific command (e.g., fire-1-new)
Test Suites
E2E Tests (8 tests)
End-to-end tests verify each command works in isolation:
| Test | Command | Verification |
|---|---|---|
| 1 | /fire-1a-new |
Creates correct .planning/ structure with all required files |
| 2 | /fire-2-plan |
Generates valid BLUEPRINT.md with skills references and must-haves |
| 3 | /fire-3-execute |
Completes breath execution with RECORD.md output |
| 4 | /fire-4-verify |
Runs all Must-Haves + WARRIOR validation checks |
| 5 | /fire-5-handoff |
Creates unified fire-handoff.md with 7-step format |
| 6 | /fire-6-resume |
Restores full context from handoff file |
| 7 | /fire-search |
Finds relevant skills from library by query |
| 8 | /fire-contribute |
Adds new skill to library with proper structure |
Integration Tests (3 tests)
Integration tests verify cross-component functionality:
| Test | Scenario | Verification |
|---|---|---|
| 1 | Full Workflow | Complete cycle: new -> plan -> execute -> verify -> handoff -> resume |
| 2 | Skills Sync | Bidirectional sync between project and global skills library |
| 3 | Hooks | SessionStart hook fires and injects context correctly |
Test Execution Process
Step 1: Parse Arguments
├── Check for --e2e, --integration, or --command flags
├── Determine which test suite(s) to run
└── Validate command name if --command specified
Step 2: Create Temp Test Directory
├── Create isolated test directory: $TEMP/dominion-flow-test-{timestamp}/
├── Initialize minimal project structure
├── Copy test fixtures from tests/fixtures/
└── Set up clean environment variables
Step 3: Run Selected Test Suite(s)
├── E2E Tests:
│ ├── For each command test:
│ │ ├── Execute command in test directory
│ │ ├── Verify expected outputs exist
│ │ ├── Validate file contents/structure
│ │ └── Record pass/fail status
│ └── Aggregate E2E results
│
├── Integration Tests:
│ ├── Full workflow test:
│ │ ├── Run complete workflow sequence
│ │ ├── Verify state transitions
│ │ └── Check final artifacts
│ ├── Skills sync test:
│ │ ├── Create test skill locally
│ │ ├── Sync to global library
│ │ ├── Verify bidirectional sync
│ │ └── Clean up test skills
│ └── Hooks test:
│ ├── Trigger SessionStart event
│ ├── Verify context injection
│ └── Check hook output
│
└── Command-specific test (if --command used):
├── Run only the specified command test
└── Provide detailed output
Step 4: Capture Results
├── Collect pass/fail for each test
├── Capture any error messages
├── Record execution times
└── Note any warnings or skipped tests
Step 5: Generate Test Report
├── Create formatted report (see format below)
├── Calculate totals and percentages
├── Highlight failures with details
└── Provide recommendations for fixes
Step 6: Clean Up
├── Remove temp test directory
├── Restore any modified global state
└── Clear test environment variables
Step 7: Return Status
├── Exit 0 if all tests passed
├── Exit 1 if any tests failed
└── Display summary message
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.
- yesterday First seen · 521 lines · 17 tokens per session scan C 2509ccac7d11
fire-test is a command published in the GitHub repository ThierryN/fire-flow (77 stars, last pushed 22d ago), licensed MIT. It adds 17 tokens to every session and 4,306 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
e2e
使用 Playwright 生成并运行端到端 (E2E) 测试.
explore-release
Model tier: opus — Opus session for the orchestrator. Charter agents run as parallel general-purpose background Tasks; tier them by cost (sonnet is usually sufficient for a single charter's execution).
t00-playwright-test
Playwright UI 测试 — 在真实浏览器中测试本地运行的 Web App,验证登录流程、功能路径、边缘案例。PJ90 等前端项目适用。.
full-test
Run the full end-to-end validation suite to ensure nothing is broken across the entire project.
tests
Test coverage quality analysis — behavioral coverage, critical gap identification, test resilience evaluation.
test
Goal: Validate implementation from user perspective.