Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/humanerd-drew/opencode-drewgentnpx agentmods add skills/humanerd-drew/opencode-drewgent/cf-workers-integrationWrote 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/skills/humanerd-drew/opencode-drewgent/cf-workers-integration)<a href="https://agentmods.dev/skills/humanerd-drew/opencode-drewgent/cf-workers-integration"><img src="https://agentmods.dev/badge/skills/humanerd-drew/opencode-drewgent/cf-workers-integration.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 5.1 | $0.00036 | $0.03415 |
| Opus 5 | $0.00018 | $0.01707 |
| Sonnet 5 | $0.00007 | $0.00683 |
| Haiku 4.5 | $0.00004 | $0.00342 |
Grade C, and why
cf-workers-integration scanned grade C with 2 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 3d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **SQLite lock**: `rm -rf .wrangler/state/v3/d1/` and `pkill -f wrangler` to clear port/lock on crash. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
export default { async fetch(request, env, ctx) { How it starts
The opening of the file, as written. The whole thing — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cloudflare Workers Project Integration
Integrating separate codebases (NAS controllers, analysis engines, LLM callers) into one coherent CF Workers project.
Trigger
- Multiple codebases need merging (NAS + working copy + external)
- Worker.ts monolithic → router + controllers refactoring
- LLM caller consolidation (duplicated DeepSeek/NVIDIA fallback)
- Data source unification (master JSON for shared constants)
Step 1 — Map structure
diff -rq src/ /source/path/src/ — find unique files on each side.
Step 2 — Find dead code
Search for orphaned files (imported 0 times), duplicated utilities, hardcoded constants.
Step 3 — Shared infra modules
src/data/<domain>-constants.json— master constantssrc/utils/llm.ts— unified DeepSeek + NVIDIA callerssrc/analysis/types.ts— analysis report types
Step 4 — Worker.ts → router pattern
import { handleX } from './src/controllers/x';
export default { async fetch(request, env, ctx) {
if (url.pathname === '/api/x') return handleX(request, env, url, ctx);
return env.ASSETS.fetch(request);
}};
Step 5 — LLM unification
src/utils/llm.ts provides:
callDeepSeek(env, systemPrompt, userContent, maxTokens)— direct DeepSeek callcallNvidiaWithFallback(env, systemPrompt, userContent, maxTokens)— 3-key NVIDIA NIM + 28s per-key timeout + AbortControllercallLLMJson(env, systemPrompt, userContent)— DeepSeek→NVIDIA fallback chain; throws on all failureextractJsonObject(text)— regex-based JSON extraction from LLM response
All controllers import from this single module instead of duplicating API call logic.
Step 6 — API Response Enrichment (injectAnalysisReport pattern)
Intercept an API response, add computed data, return enhanced response without breaking existing consumers:
async function injectAnalysisReport(response: Response): Promise<Response> {
const clone = response.clone();
const body = await clone.json();
body.data.analysisReport = analyze(body.data);
return new Response(JSON.stringify(body), {
status: response.status,
headers: response.headers, // preserve CORS/auth headers
});
}
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 3d ago First seen · 239 lines · 36 tokens per session scan C 40a9079c6b9d
cf-workers-integration is a skill published in the GitHub repository humanerd-drew/opencode-drewgent (2 stars, last pushed 1mo ago), licensed MIT. It adds 36 tokens to every session and 3,415 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
stripe-projects
Provision SaaS services + sync creds via Stripe Projects.
architecture-decision-record
ADR templates in the Nygard format with context, decision, consequences, and alternatives. Use when writing ADRs, recording an architectural decision, or evaluating options.
architecture-paradigm-microservices
Applies microservices for independent deployment and per-service scaling. Use when teams need autonomous release cycles with distinct capability scaling needs.
multi-tenant-architecture
Designs tenant isolation, hostname routing, custom-domain lifecycle, and plan limits on Cloudflare or Vercel. Use when asked to "isolate tenant data", "support custom domains", "build a white-label platform", or assess PSL registration. For general module structure use codebase-architecture; for SEO content use…
architecture-paradigm-serverless
Applies serverless FaaS patterns for event-driven workloads. Use when designing bursty workloads with minimal infrastructure and pay-per-execution cost model.
microservices-expert
Expert-level microservices architecture, patterns, service mesh, and distributed systems. Use when the user mentions distributed systems, service mesh, or architecture, or when the task involves Microservices Principles, Architecture Patterns, Communication, or Data Management.