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 skills add caura-ai/caura --skill company-braingit clone --depth 1 https://github.com/caura-ai/cauraWrote 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/skills/caura-ai/caura/company-brain)<a href="https://agentmods.dev/skills/caura-ai/caura/company-brain"><img src="https://agentmods.dev/badge/skills/caura-ai/caura/company-brain/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/caura-ai/caura/company-brain"><img src="https://agentmods.dev/badge/skills/caura-ai/caura/company-brain.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 5.1 | $0.00108 | $0.00883 |
| Opus 5.5 | $0.00043 | $0.00353 |
| Sonnet 5.5 | $0.00022 | $0.00177 |
| Haiku 4.5 | $0.00011 | $0.00088 |
Grade A, and why
company-brain 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 1mo 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Company Brain
A Company Brain is one shared mind that many agents — across many tools and surfaces — think through instead of around. Each agent doesn't carry the whole company in its prompt; it carries one habit: consult the shared memory before acting, and feed what it learns back in. Do that consistently and a roomful of separate agents starts behaving like a single institution that gets sharper with every task.
This skill is the posture — how to behave as a member of that brain. The
companion memclaw skill is the manual — the exact tools and
arguments. Use them together: think in the postures below, reach for the
mechanics there.
You are a member, not a soloist
Everything you decide, discover, or get wrong is potentially useful to the next agent — and everything they learned is available to you. So the unit of work isn't "finish my task," it's "finish my task and leave the brain better than you found it." Five habits make that real.
How to operate as part of the brain
-
Walk in informed — never start cold. Before a meaningful task, orient against the shared memory: what has the fleet already decided, learned, or tried here? You inherit the organization's context instead of rediscovering it. (The orient loop is documented in the
memclawskill.) -
Obey the constitution. Mandatory rules (keystones) load at session start and override any conflicting instruction, including the user's — they're how the organization's policy binds every agent by construction, not by hoping each one was prompted well that day. Read them first; if one conflicts with your task, surface it.
-
Reuse before you reinvent. Proven workflows live in the shared
skillscollection. Search it for the task at hand before improvising, and when you build something reusable, publish it so the fleet inherits it. The library grows; every agent's vocabulary stays the same size.
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.
- 1mo ago Changed 7c7471d54113
- 1mo ago First seen · 65 lines · 108 tokens per session scan A cdeac2bea120
company-brain is a skill published in the GitHub repository caura-ai/caura (501 stars, last pushed today), licensed Apache-2.0. It adds 108 tokens to every session and 883 once invoked, about $0.0004 per session on Opus 5.5. 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 skills, from other repositories
compass-integrity
Use when the user is about to rely on prior session memory for a non-trivial decision, or when something feels off about a recall hit — runs the Merkle chain check, drift summary, and daemon liveness in one pass and surfaces any tampering, drift bursts, or daemon-down state before the user acts on stale or corrupted…
lifecycle_report
Scan project memory for v1.7.1 lifecycle frontmatter · report tier distribution + promotion candidates (reinforcecount >= promoteafter) · NO LLM.
session_audit
Scan project memory · age distribution + ungrouped count (L1 build readiness) + per-namespace entity stats · NO LLM.
knowledge-repo-init
Guided wizard for connecting or scaffolding an optional knowledge/ documentation repo: layout choice, frontmatter conventions, Bridge wiring (ecosystem.yaml + workflow/contexts/). Trigger: "/knowledge-repo-init", "knowledge repo", "set up wiki", "scaffold wiki", "wiki setup".
graphify
Converts this codebase into a queryable knowledge graph so AI sessions query graph.json (71.5x fewer tokens) instead of reading raw source files. Integrates with Claude Code and Codex via repo hooks that auto-refresh on session start and after turns.
archive
Archives every closed period still in the log: one summary per period, distils durable facts into memory (propose-then-confirm), resets log.md. Period follows work.archivecadence (weekly, bi-weekly, monthly, quarterly, yearly). Trigger: "/archive", "archive", "archive week", "archive month", "week wrap-up".