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.
git clone --depth 1 https://github.com/alifanov/darkflowWrote 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/alifanov/darkflow/seo-check)<a href="https://agentmods.dev/commands/alifanov/darkflow/seo-check"><img src="https://agentmods.dev/badge/commands/alifanov/darkflow/seo-check/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/alifanov/darkflow/seo-check"><img src="https://agentmods.dev/badge/commands/alifanov/darkflow/seo-check.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00022 | $0.07254 |
| Opus 5.5 | $0.00009 | $0.02902 |
| Sonnet 5.5 | $0.00004 | $0.01451 |
| Haiku 4.5 | $0.00002 | $0.00725 |
Grade A, and why
seo-check scanned grade A 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| `submitted` | URL count **as of that fetch** — compare with a live `curl` of the sitemap to spot drift | How it starts
The opening of the file, as written. The whole thing — 407 lines — stays where its author put it; the contents beside it link to each section on GitHub.
This routine has two halves:
- GSC data — what's actually happening in search (positions, CTR, impressions, indexing).
- SEO audit — why it's happening (on-page + technical issues in the codebase / live pages).
GSC is the only half that carries real measured data — everything else is a static read of the code. When the GSC MCP is connected, Step 2 is never optional and never a footnote: run it first and let its numbers set the priority of what Step 3 finds. Skipping it because "the audit already found things" inverts the routine.
Step 1 — Read project config
Load the project config (contract in .darkflow.d/claude.md → Project config). Uses: language, domain. If domain is absent, try to auto-discover the production URL (Coolify FQDN, vercel.json/.vercel/project.json, netlify.toml, CNAME); if none is found, skip the live-page checks and audit the codebase only.
Step 2 — GSC data analysis
IMPORTANT: Never use the browser to access Google Search Console. All GSC data must be fetched exclusively via the Google Search Console MCP tools. If the GSC MCP is not connected or its tools are unavailable, skip this step (do NOT abort the whole routine — Step 3 still runs) and note in the snapshot:
GSC data skipped: Google Search Console MCP is not connected.
Connect the GSC MCP in your project's .claude/settings.json to enable it.
Do not fall back to browser automation for GSC data.
Check Google Search Console data for the last week using MCP tools. Analyse positions, CTR, impressions, and indexing issues. For each meaningful finding, suggest a concrete fix and file it as a task (see task format below) with --source gsc.
Look in the repo before asking for GSC data. Projects often keep exports of Performance
reports in-tree — grep -rl "impressions" docs/insights docs/logs 2>/dev/null or a glob for
docs/insights/search-console/*.md. Those tables carry per-query and per-page history the API
window no longer covers, and reading them costs nothing. Found in four naturalwrite files only
after a full round of "please export this from GSC" — check first.
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.
- 1mo ago First seen · 407 lines · 22 tokens per session scan A 82d646bafa2f
seo-check is a command published in the GitHub repository alifanov/darkflow (4 stars, last pushed yesterday), licensed MIT. It adds 22 tokens to every session and 7,254 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
local
Command "local" from getsentry/toolkit, covering examples, start the server and tail events (default), run your app with the local server auto-enabled, use a custom port and only show errors and logs (filter out transactions).
auth
Command "auth" from getsentry/toolkit, covering examples, oauth login (recommended), token login, read-only oauth login and custom oauth scopes.
debug-files
Command "debug-files" from getsentry/toolkit, covering examples, inspect a debug information file (auto-detects the format), bundle jvm sources with a debug id, exclude additional directories and output as json.
feedback
Modern User Feedback is stored as Feedback issues. The command always limits issue searches to issue.category:feedback; it does not use the legacy User Reports API.
api
Endpoints are relative to /api/0/ — the prefix is added automatically. Absolute HTTP(S) Sentry URLs also work; their origin is validated against your authenticated host.
init
Authentication: Interactive runs start the OAuth login flow automatically when credentials are missing or the stored session has expired. Non-interactive and CI runs must authenticate before running sentry init.