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/adilkalam/orca/iosgit clone --depth 1 https://github.com/adilkalam/orcaWhat 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 | $0.00014 | $0.06208 |
| Opus 5 | $0.00007 | $0.03104 |
| Sonnet 5 | $0.00003 | $0.01242 |
| Haiku 4.5 | $0.00001 | $0.00621 |
Grade A, and why
ios 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 2d 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 — 693 lines — stays where its author put it; the contents beside it link to each section on GitHub.
STOP - DELEGATION ONLY
Before you do ANYTHING else, read this.
This slash command EXISTS to delegate work to agents. Not to do work directly.
NEVER acceptable:
- "This is simple, I'll just do it directly"
- "Let me quickly fix this"
- "I can handle this without agents"
- Using Edit/Write tools to make changes yourself
ALWAYS required:
- Parse the arguments
- Determine routing (--tweak, default, --complex)
- Spawn specialists single-level via the
Agenttool (you are the orchestrator, in the main thread — no orchestrator subagent; OS 7.1)
Even --tweak delegates to a builder. It skips gates, not agents.
If you are about to Edit/Write, STOP. Delegate instead.
/ios - iOS Lane Orchestrator (OS 7.1)
Use this command for native iOS work (Swift/SwiftUI/UIKit, Xcode, device features).
Usage
/ios "add haptic feedback to save button" # Default: light path + design gates
/ios --tweak "fix button padding" # Fast: light path, no gates
/ios --complex "multi-screen feature" # Full: architect + builder + all gates
/ios "implement requirement 2025-11-25-0930-workspace-sharing" # Full path with spec
CRITICAL ROLE BOUNDARY
YOU ARE AN ORCHESTRATOR. YOU NEVER WRITE CODE.
- DO NOT use Edit/Write tools
- DO NOT bypass the agent system
- DELEGATE via the
Agenttool only (single-level, from the main thread) - Update phase_state.json to track progress
- Resume from interruptions without abandoning pipeline
0. Parse Arguments & Detect Mode
Check for flags:
$ARGUMENTS contains "--light" → Section 2.1 (Light Orchestrator, NO confirmation)
$ARGUMENTS contains "--tweak" → Section 2.2 (Builder Direct, NO confirmation)
$ARGUMENTS contains "--complex" → Section 3 (Full Pipeline with confirmation)
No flag → Section 3 (Light Orchestrator WITH confirmation)
0.1 Recording Context (OS 7.1)
Session activity is captured automatically by orca-record hooks. Before delegating to agents, inject prior session context for continuity.
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.
- 2d ago First seen · 693 lines · 14 tokens per session scan A 390816d23cc0
ios is a command published in the GitHub repository adilkalam/orca (2 stars, last pushed 1mo ago), licensed MIT. It adds 14 tokens to every session and 6,208 once invoked, about $0.0001 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 2026-08-31.
Other commands, from other repositories
implement
yes, thank you! please proceed with implementation.
suggest-solution
if there is better solution, please suggest it.
android
Command "android" from zhagqn/agentwork, covering /android [subcommand] [args] and 执行前必读.
compose-test
Run Compose UI tests with Espresso. Verify critical user flows.
app-store-audit
Run an enterprise pre submission compliance audit on an iOS or Android app against Apple App Store and Google Play rejection rules. Pass a project path or run from the project root.
build-verify
Run the full build / test / analyze / format verification pass for this Flutter app.