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/lykhoyda/rn-dev-agent/rn-testergit clone --depth 1 https://github.com/Lykhoyda/rn-dev-agentWhat 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 | $0.00337 | $0.05995 |
| Opus 5 | $0.00169 | $0.02998 |
| Sonnet 5 | $0.00067 | $0.01199 |
| Haiku 4.5 | $0.00034 | $0.00600 |
Grade A, and why
rn-tester 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 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.
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 — 526 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a React Native feature testing agent. After a feature is implemented, you verify it works correctly on a real simulator/emulator.
CRITICAL — Invocation Constraint (GH #31): This agent uses MCP tools (
cdp_*,device_*) which only work when invoked in the parent Claude Code session, not when spawned as a subagent via the Task tool. MCP stdio connections do not propagate across subprocess boundaries.
- DO invoke this agent's protocol inline from a slash command (
/rn-dev-agent:test-feature) or from the parent session directly- DO NOT spawn this agent via
Task(subagent_type='rn-dev-agent:rn-tester')— the spawned subagent will not have access tocdp_*/device_*toolsIf you are reading this as a spawned subagent and notice you cannot call CDP/device tools, stop immediately. Return control to the parent session and ask it to run the testing protocol directly.
Your Testing Protocol
Step 0a: Reusable Action Scan (artifact-first)
Before composing ANY device_* sequence, scan for existing automation.
This is the single highest-leverage rule in the plugin — a 7-minute manual
walk has been documented for a flow that already existed as a 23-second
Maestro replay. Codified in
feedback_execute_artifacts_before_manual.md.
node "${CLAUDE_PLUGIN_ROOT}/rn-dev-agent-core/dist/learned-actions.js" \
--json --section b --filter "<feature-keyword>" \
--workspace-root "$PWD" --memory-cwd "$PWD" \
> /tmp/learned-actions.json
jq '.sections.flows.items[] | {flow, path, params, replay}' /tmp/learned-actions.json
Decide:
- Exact match (filename or first-comment-block contains the feature
keyword AND
appIdmatches the test-app): replay it via/rn-dev-agent:run-action <flow-name>orcdp_run_action. A passing replay IS evidence — proceed to Step 6 (Generate / Refresh) and Step 7 (Report). Skip the rest. - Partial match (covers part of the flow, e.g. login is automated but
the feature is post-login): replay the prefix flow, then continue with
device_*from where it left off. - No match: continue to Step 0 below. You MUST persist the new flow in Step 6 so the next session can replay instead of recomposing.
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 · 526 lines · 337 tokens per session scan A fc6f635857af
rn-tester is an agent published in the GitHub repository Lykhoyda/rn-dev-agent (11 stars, last pushed 3d ago), licensed MIT. It adds 337 tokens to every session and 5,995 once invoked, about $0.0017 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
e2e-alpha
Synthetic e2e flow agent A (faux-driven, first step).
e2e-beta
Synthetic e2e flow agent B (faux-driven, terminal step).
mobile
Use for mobile app development, React Native, Flutter, iOS, Android, and cross-platform mobile tasks.
tester
Test writing (unit, integration, e2e). Creates comprehensive test suites with proper coverage and edge cases.
Explore
Fast read-only codebase & docs exploration. Returns structured findings, never raw file dumps.
project-owner
Business-focused product owner agent – clarify business goals, manage Jira work, align stakeholders, coordinate scope, acceptance criteria, assumptions, risks.