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/lykhoyda/rn-dev-agent/build-and-testgit clone --depth 1 https://github.com/Lykhoyda/rn-dev-agentWrote 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/lykhoyda/rn-dev-agent/build-and-test)<a href="https://agentmods.dev/commands/lykhoyda/rn-dev-agent/build-and-test"><img src="https://agentmods.dev/badge/commands/lykhoyda/rn-dev-agent/build-and-test.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 | $0.00031 | $0.00982 |
| Opus 5 | $0.00015 | $0.00491 |
| Sonnet 5 | $0.00006 | $0.00196 |
| Haiku 4.5 | $0.00003 | $0.00098 |
Grade C, and why
build-and-test scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
`trap 'rm -rf -- "$artifact_dir"' EXIT` in that same scope. Then run How it starts
The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Build the app and test this feature: $ARGUMENTS
Run the build + test protocol INLINE (parent session)
Important (GH #31): Do NOT spawn the
rn-testeragent via the Task tool. MCP tools (cdp_*,device_*) are not available in spawned subagents. Execute the build + test protocol directly in this parent session.
Phase A — Build pre-flight (run in this session):
- Resolve session authority — call
rn_session(action="status")and require the intended worktree, platform, exact UUID/serial, and app ID.device_listmay diagnose available hardware but never selects the target. - Check if app is already running — call passive
cdp_status. If the session's exact signed target is connected andapp.dev == true, skip to Phase B. - Build / install (substitute
<platform>withiosorandroid):- No
--easflag: after previewing and confirming the session integration, run literalpnpm iosorpnpm android. The adapter injects the exact device and allocated Metro port and records the build/install receipt. For Expo Android, continue to bind and report the exact adb serial: the adapter uniquely translates it to Expo's model/AVD display name only for--device, pins adb withANDROID_SERIAL, and refusesEXPO_DEVICE_IDENTITY_MISMATCHbefore Expo on a missing, unauthorized, duplicate, foreign, or drifted mapping. - With
--easflag: enter one shell scope, create<artifact-dir>withartifact_dir=$(mktemp -d), and immediately registertrap 'rm -rf -- "$artifact_dir"' EXITin that same scope. Then runbash "$CLAUDE_PLUGIN_ROOT/scripts/eas_resolve_artifact.sh" "<platform>" "<profile>" "<artifact-dir>", parse a successful absolute artifact path, and runbash "$CLAUDE_PLUGIN_ROOT/scripts/expo_ensure_running.sh" "<platform>" --device-id "<device-id>" --artifact "<path>". Keep resolution and installation inside the trapped scope so every success or failure path cleans only that exact caller-owned directory after the install attempt.
- No
- Start Metro only through the literal integrated package script; an ambient default-port Metro is diagnostic, not authoritative.
- Confirm CDP — call
cdp_connectfor the exact signed target, then passivecdp_status.
Phase B — Run the rn-tester protocol (load rn-testing skill):
Follow the same protocol as /rn-dev-agent:test-feature, INCLUDING its
mandatory Step 0 — scan saved actions (/rn-dev-agent:list-learned-actions)
and replay a matching flow BEFORE composing any device_* primitives — then
environment check, understand the feature, plan, navigate, execute+verify,
edge cases, generate persistent test.
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.
- 4d ago First seen · 77 lines · 31 tokens per session scan C 70f44f0cec2b
build-and-test is a command published in the GitHub repository Lykhoyda/rn-dev-agent (11 stars, last pushed today), licensed MIT. It adds 31 tokens to every session and 982 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
dashboard-flow-auto
Toggle autonomous mode for a session's flow. Usage /dashboard:flow-auto.
dashboard-git-branches
List git branches for the current dir (current marked ). Runs locally, no LLM.
brainstorm
Command dispatcher for brainstorm.
propose
Command dispatcher for propose.
accept
Command dispatcher for accept.
apply
Command dispatcher for apply.