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 instructions/comisai/comis/agents-mdgit clone --depth 1 https://github.com/comisai/comisWhat 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 | $0.13563 | $0.13563 |
| Opus 5 | $0.06782 | $0.06782 |
| Sonnet 5 | $0.02713 | $0.02713 |
| Haiku 4.5 | $0.01356 | $0.01356 |
Grade A, and why
comis AGENTS.md 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.
This is a copy
86% identical to gomaa AGENTS.md — 455 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 408 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Comis Engineering Protocol
Default working protocol for coding agents. Scope: entire repository.
1) Architecture
Hexagonal (ports + adapters). Core defines port interfaces; adapters implement them; everything is wired via AppContainer in packages/core/src/bootstrap.ts (composition root). Extending Comis means implementing a port interface and wiring it in bootstrap — not cross-cutting rewrites.
Extension points in packages/core/src/:
ports/— port interfaces (*Portsuffix):ChannelPort,ChannelPluginPort,MemoryPort,SkillPort,EmbeddingPort,MediaResolverPort,TranscriptionPort,TTSPort,ImageAnalysisPort,VisionPort,FileExtractionPort,OutputGuardPort,SecretStorePort,DeviceIdentityPort,CredentialMappingPort,PluginPort,DeliveryQueuePort,DeliveryMirrorPort,ClockPort,EnvPort,TimerPort(+TimerHandle),ContextStorePort,SessionStorePort,FileLockPort,CapabilityServiceControlPort,ManagedRunStorePort,ManagedRunContentPort,WorkspaceLeasePort,ExecutionAttachmentPort, hook types. Adapters live in the consumer package (e.g.,@comis/memoryfor store ports) or@comis/infrafor runtime adapters (createSystemClock,createSystemEnv,createSystemTimers).runtime/— sanctioned-root in-package helpers (system-time.ts:systemNowMs,systemNowDate,systemDateFrom,systemSleep,systemSetTimeout/systemClearTimeout,systemSetInterval/systemClearInterval,systemScheduleTimeout,systemGetEnv,systemEnvSnapshot). Use these only at trust-boundary call sites that genuinely cannot accept an injected port (top-level loggers, OAuth poll loops). Secrets must still go throughSecretManager.domain/— Zod-validated domain types (NormalizedMessage,MemoryEntry,AgentResponse,ExecutionGraph,SubagentResult,ApprovalRequest,CredentialMapping,SecretRef, etc.). Define schema → infer type withz.infer.security/— security primitives:safePath,validateUrl(SSRF),SecretManager,SecretsCrypto(AES-256-GCM),ScopedSecretManager,SecretRefResolver,ActionClassifier,AuditAggregator,InputSecurityGuard,validateInput,OutputGuard,MemoryWriteValidator,wrapExternalContent,sanitizeLogString,CanaryToken, injection patterns + rate limiter.config/— Zod schemas, layered config (defaults → YAML files → env overrides). Paths viaCOMIS_CONFIG_PATHS(comma-separated). Runtime writes useconfig.patch/config.apply, while immutable topology is changed in config files and applied on restart;docs/reference/config-yaml.mdxowns the field contracts.event-bus/—TypedEventBuswith strongly-typed event families includingAgentEvents,ChannelEvents,MessagingEvents,InfraEvents,CapabilityServiceEvents, andManagedRunEvents. Logging supplements events, does not replace them.hooks/—PluginRegistry+HookRunnerfor plugin lifecycle.context/— AsyncLocalStorage request-scoped context viarunWithContext()/getContext().bootstrap.ts— composition root →AppContainer.
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 · 408 lines · 13,563 tokens per session scan A e1851f207862
comis AGENTS.md is an instructions file published in the GitHub repository comisai/comis (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 13,563 tokens to every session, about $0.0678 per session on Opus 5. A static security scan graded it A with 0 findings. It is 86% identical to gomaa AGENTS.md, differing in 455 lines, and is treated as a copy.
Other instructions, from other repositories
PuPu AGENTS.md
Instructions for haoxiang-xu/PuPu, covering gitnexus — code intelligence, always do, never do, resources and cli.
PuPu CLAUDE.md
Instructions for haoxiang-xu/PuPu, covering gitnexus — code intelligence, always do, never do, resources and cli.
takt AGENTS.md
Instructions for nrslib/takt, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
commonly CLAUDE.md
Instructions for Team-Commonly/commonly, covering claude.md / agents.md, 🧠 product vision & architecture philosophy, what commonly is, the architecture model and key concepts.
VaultysClaw agent-controller.instructions.md
Use when working on the agent-controller package: tools, skills, agent runtime, WebSocket client, local SQLite DB, LLM integration, or CLI modes.
open-ace CLAUDE.md
Instructions for open-ace/open-ace, covering claude.md, where process documents go, pushing, test placement and ci semantics and schema snapshots (schema-sync ci).