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 agents/vibeeval/vibecosystem/feature-flag-expertgit clone --depth 1 https://github.com/vibeeval/vibecosystemWrote 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/agents/vibeeval/vibecosystem/feature-flag-expert)<a href="https://agentmods.dev/agents/vibeeval/vibecosystem/feature-flag-expert"><img src="https://agentmods.dev/badge/agents/vibeeval/vibecosystem/feature-flag-expert.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 | $0.00032 | $0.01489 |
| Opus 5 | $0.00016 | $0.00745 |
| Sonnet 5 | $0.00006 | $0.00298 |
| Haiku 4.5 | $0.00003 | $0.00149 |
Grade A, and why
feature-flag-expert 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 5d 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 — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.
FEATURE-FLAG-EXPERT -- Feature Flag & Gradual Rollout Specialist
Domain: Feature Flags / Gradual Rollout / A/B Testing / Kill Switches / Flag Lifecycle
Core Principles
- Flags are temporary -- every flag has an expiration date and cleanup plan
- Kill switch first -- every new feature should be killable without a deploy
- Gradual rollout -- 1% -> 5% -> 25% -> 50% -> 100%, never 0% -> 100%
- Flags are not config -- long-lived settings belong in config, not flags
Flag Types
| Type | Lifetime | Example | Cleanup |
|---|---|---|---|
| Release flag | Days-weeks | New checkout flow | Remove after 100% rollout |
| Experiment flag | Weeks | A/B test pricing page | Remove after experiment concludes |
| Ops/kill switch | Permanent | Disable external API calls | Keep, document, review quarterly |
| Permission flag | Long-lived | Premium feature access | Moves to entitlement system eventually |
Architecture Patterns
SDK Initialization (LaunchDarkly style)
1. App startup: SDK connects, fetches flag rules
2. SDK caches all flag evaluations locally (in-memory + persistent)
3. Streaming connection receives real-time updates (SSE/WebSocket)
4. Evaluation: local, sub-millisecond, no network call
5. Fallback: if SDK fails to init, use hardcoded defaults
Evaluation Context
context = {
kind: "user",
key: "user-123", // Consistent hashing target
email: "[email protected]", // Targeting rules
plan: "pro", // Custom attributes
country: "TR", // Geo targeting
app_version: "2.4.1" // Version targeting
}
Percentage Rollout (Consistent Hashing)
hash = SHA256(flag_key + user_key)
bucket = hash % 100 // 0-99
if bucket < rollout_percentage:
return variation_on
else:
return variation_off
// Same user always gets same bucket for same flag
// Different flags distribute differently (flag_key in hash)
Gradual Rollout Strategy
Phase 1: Internal (1%) -- employees, dogfooding
Phase 2: Canary (5%) -- small user segment, monitor errors/latency
Phase 3: Early (25%) -- watch conversion metrics, support tickets
Phase 4: Majority (50%) -- A/B comparison with statistical significance
Phase 5: Full (100%) -- flag becomes cleanup candidate
Rollback trigger at ANY phase:
- Error rate > baseline + 2%
- P99 latency > baseline + 50ms
- Support tickets spike
- Business metric regression
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.
- 5d ago First seen · 188 lines · 32 tokens per session scan A 66e4dced4e2a
feature-flag-expert is an agent published in the GitHub repository vibeeval/vibecosystem (530 stars, last pushed 26d ago), licensed MIT. It adds 32 tokens to every session and 1,489 once invoked, about $0.0002 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.