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 kombohq/company-brain --skill add-connectorgit clone --depth 1 https://github.com/kombohq/company-brainWrote 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/kombohq/company-brain/add-connector)<a href="https://agentmods.dev/skills/kombohq/company-brain/add-connector"><img src="https://agentmods.dev/badge/skills/kombohq/company-brain/add-connector/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/kombohq/company-brain/add-connector"><img src="https://agentmods.dev/badge/skills/kombohq/company-brain/add-connector.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00074 | $0.01351 |
| Opus 5 | $0.00037 | $0.00675 |
| Sonnet 5 | $0.00015 | $0.00270 |
| Haiku 4.5 | $0.00007 | $0.00135 |
Grade A, and why
add-connector 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 11d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Add a connector (sync a new data source)
This repo pulls external data into context/ as plain files so an agent can grep, read, and follow links. This skill is about adding the sync code, not about answering questions from already-synced data.
Principles
- Bun only. Scripts run with
bun run …; no build step, no npm/pnpm/yarn. - Keep it lean. Do the smallest thing that works. Add a dependency only when there's a concrete need, and never drag in tooling from other repos (no databases, queues, web frameworks) just because a similar connector elsewhere had them.
- Agent-friendly output. Write Markdown with YAML frontmatter. Record the source id/url in frontmatter so files can be linked back and pruned.
- Deterministic output. Writing the same upstream state must produce byte-identical files. Never write volatile metadata (timestamps like
fetched_at, random ids); frontmatter should carry only durable identity (source id/url, title). Volatile fields create a huge diff on every run of the self-commit workflow. - Cross-link the repo. Context files should reference related files in the repo by relative path so the agent can navigate by following links. Rewrite upstream links to point at the synced files; when a target isn't synced, keep the full external URL.
- Incremental, prune only when needed. Refetch only what changed. Deletion tracking (pruning files whose source disappeared upstream) isn't always needed, consider whether this source needs it and ask the user if unsure. When you do prune, skip pruning when a run discovers nothing so a transient outage can't wipe the mirror.
- Fail loud. Most failures should crash the run so they surface in the GitHub Actions log, never silently produce a partial mirror. Only swallow a failure in incremental cases where retries are exhausted and a partial update is acceptable (follow Notion's
continue-on-error+ surface-after-commit pattern). - Bounded concurrency. For list-then-detail jobs use
processParallelfromsrc/lib/process-parallel.tswith a fixed concurrency constant beside the code. No unboundedPromise.all, no accumulating huge in-memory arrays, fetch → write → release. - Minimal credentials. Require the least-privileged token/scope, and document exactly how to create it: link to the page where the credential is configured and spell out the precise permissions needed.
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.
- 11d ago First seen · 67 lines · 74 tokens per session scan A 8fd6e11eaec7
add-connector is a skill published in the GitHub repository kombohq/company-brain (20 stars, last pushed 2mo ago), licensed MIT. It adds 74 tokens to every session and 1,351 once invoked, about $0.0004 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 skills, from other repositories
ensembl-database
Query Ensembl genome database REST API for 250+ species. Gene lookups, sequence retrieval, variant analysis, comparative genomics, orthologs, VEP predictions, for genomic research.
colab-finetuning
Fine-tune LLMs on Google Colab GPUs directly from openscience. Connects to Colab runtimes via WebSocket bridge for remote training with Unsloth. Supports SFT, GRPO, DPO, vision, and TTS workflows on free T4 to Pro A100 GPUs.
swarm-pr-review
Run a graph-guided, tool-augmented PR review using context packing, parallel exploration, mandatory repository-agnostic risk-family coverage with dispatch scaled to diff size and risk, independent reviewer validation, critic challenge, and metrics writeback. Use for deep pull request review with low false-positive…
loop
Full execution protocol for MODE: LOOP — the compound-engineering loop: brainstorm → plan → build → review → improve, iterating under defense-in-depth stop conditions with generator/critic separation, durable resumable state, and mandatory compounding learning capture. Loaded on demand by the architect when the loop…
restore-internals-seams-in-finally-blocks-after-each-test
When delegating a task affected by this skill, include.
google-meet
Google Meet via gws: create spaces, fetch join links, list recordings.