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/dvnghiem/flowdeck/fd-verifygit clone --depth 1 https://github.com/DVNghiem/FlowDeckWhat 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 | $0.00025 | $0.01408 |
| Opus 5 | $0.00013 | $0.00704 |
| Sonnet 5 | $0.00005 | $0.00282 |
| Haiku 4.5 | $0.00003 | $0.00141 |
Grade A, and why
fd-verify 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.
How it starts
The opening of the file, as written. The whole thing — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify
Run the verification pipeline for the active topic. /fd-done will not close a task
until this passes.
Input: $ARGUMENTS — optional --topic=<slug> to target a topic other than the
active one.
Pre-flight
- Check
~/.fd-plan/<slug>/STATE.mdexists — if not:"No planning workspace. Run /fd-task first." - Resolve
<topic>from--topic, else fromtopicin STATE.md. - Confirm
steps_completeis non-empty — if empty, warn:"No steps completed yet. Run /fd-execute first."
Step 1: Run tests
Run the project's full test suite. Detect the runner from the project manifest —
npm test / bun test for Node, cargo test for Rust, pytest for Python, go test ./...
for Go.
All tests must pass. No failures, no unexplained skips.
If no test runner is configured, the agent MUST report that explicitly. The agent MUST NOT
pass silently: "No test script found — cannot verify. Configure a test runner before /fd-done."
Step 2: Browser tests for web projects
Detect a web project by the presence of a frontend framework in the manifest
(react, vue, svelte, next, astro) or a public/index.html.
- Playwright is configured (a
playwright.config.*exists) → run the E2E suite and fold the result into the verdict. - Web project without Playwright → suggest adding it, and note that UI behavior is unverified in this run. The agent MUST NOT fail the verdict on its absence.
- Not a web project → skip this step entirely.
Step 3: Regression check on affected files
Read ~/.fd-plan/<slug>/<topic>/affect.md and take the Affected Files list as the
regression scope.
For each affected file:
- Confirm it has test coverage. Uncovered changed files are a HIGH finding.
- Use
fdx-impactto find dependents not listed inaffect.md, and verify their tests still pass.
Log any dependent that affect.md missed — that is a planning gap worth recording.
Step 4: Review and scan
Code review (@reviewer) — over the files changed since the task began:
- Security: secrets, injection, auth gaps
- Quality: critical bugs, missing error handling, TDD discipline
- Conventions: naming, patterns, import style
- Test coverage ≥ 80% for changed files — below that is a HIGH finding
- Any
override_logentries from/fd-execute --override
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.
- 2d ago First seen · 166 lines · 25 tokens per session scan A 876b141755da
fd-verify is a command published in the GitHub repository DVNghiem/FlowDeck (24 stars, last pushed 14d ago), licensed MIT. It adds 25 tokens to every session and 1,408 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.
Other commands, from other repositories
playwright
Write or update Playwright E2E tests following project conventions.
e2e
Create and run E2E tests across web, mobile, and desktop with auto-detection and CLI-first toolchains.
frontend-e2e
Perform a thorough live E2E test of a frontend web application using browser-bridge MCP tools. Tests every page, interactive element, form, navigation flow, responsive breakpoint, and accessibility rule through real browser interaction.
auto-canary
배포 검증 — build, E2E, 브라우저 건강 검진을 실행합니다.
annotate-plan
Draft a plan and open it in the annotator UI for user review.
test-feature
Test a React Native feature on the running simulator/emulator. Verifies UI, user flows, and internal state. Generates a persistent Maestro test file.