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 skills/despotb/handelsregister-cli/handelsregisternpx skills add DespotB/handelsregister-cli --skill handelsregistergit clone --depth 1 https://github.com/DespotB/handelsregister-cliWrote 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/despotb/handelsregister-cli/handelsregister)<a href="https://agentmods.dev/skills/despotb/handelsregister-cli/handelsregister"><img src="https://agentmods.dev/badge/skills/despotb/handelsregister-cli/handelsregister.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 | $0.00125 | $0.00859 |
| Opus 5 | $0.00063 | $0.00430 |
| Sonnet 5 | $0.00025 | $0.00172 |
| Haiku 4.5 | $0.00013 | $0.00086 |
Grade A, and why
handelsregister 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 3d 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Handelsregister document fetch
Fetch official registry documents for a German company from handelsregister.de
using the hreg CLI (from the same repository as this plugin).
Steps
-
Ensure the CLI is available. Run
hreg --version. If missing, install it with the first of these tools that exists on the machine:uv tool install git+https://github.com/DespotB/handelsregister-clipipx install git+https://github.com/DespotB/handelsregister-clipip install --user git+https://github.com/DespotB/handelsregister-cli
If
hregis still not found afterwards, the install bin directory is likely not on PATH — try~/.local/bin/hreg --versionand call it by that path. If all installers fail, tell the user what to install and stop. -
Search first, never guess. Run:
hreg search "<company keywords>" --json- No results: tell the user, suggest broader keywords or
--mode any. - Several plausible results: show name + court + register number and ask which one is meant, unless the user already gave a register number.
- No results: tell the user, suggest broader keywords or
-
Fetch. Default is everything; honour a user's narrower request via
--docs:hreg fetch "<keywords>" --register "<HRB/HRA number>" -o "<outdir>" --json- Output directory: whatever the user or the project's conventions dictate;
otherwise
./<company name>/in the working directory. - The command is slow by design (about 3 seconds between requests, a full
fetch can take a few minutes). Do not lower
--delaybelow 3 and do not loop over many companies — the portal rate-limits at roughly 60 requests/hour.
- Output directory: whatever the user or the project's conventions dictate;
otherwise
-
Report. Read
manifest.jsonfrom the output directory and summarise: which documents were downloaded (highlight Gesellschafterlisten and Gesellschaftsverträge with their dates), what was skipped and why, and where the files are. Mention that duplicates were removed by checksum.
Announcements and watching
- For "what was announced recently" questions use
hreg announcements --days N [--land BE] [--category …] [--keyword …] --json. Categories: Löschungsankündigung, UmwG announcements, new document filings, and two "other" buckets.--detailsadds the full text (one extra request per item — keep the result set small first). - There is no feed of newly registered companies — since the DiRUG reform
(Aug 2022) Neueintragungen are not announced anywhere. Say so instead of
improvising. The closest free approximation is
hreg watch --search "<name keyword>"on a schedule, which reports register numbers not seen on a previous run (first run only seeds its state file).
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.
- 3d ago First seen · 63 lines · 125 tokens per session scan A 532a0c44d989
handelsregister is a skill published in the GitHub repository DespotB/handelsregister-cli (2 stars, last pushed 10d ago), licensed MIT. It adds 125 tokens to every session and 859 once invoked, about $0.0006 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 skills, from other repositories
octopus-security-audit
OWASP compliance, vulnerability scanning, and adversarial red team testing — use for security reviews.
researchers-legal
Researches court documents, indictments, plea agreements, and sentencing records. Use when the album subject involves legal proceedings or criminal cases.
ai-ml-governance
Governs models and AI systems in production — intended use, evaluation, monitoring, human oversight, documentation, and the decision to deploy or retire. Use this before deploying a model or AI feature, when defining evaluation criteria, when a model's behavior has drifted, when assessing AI risk or regulatory…
prior-art-search
Systematic 7-step methodology for comprehensive patent prior art searches and patentability assessments using BigQuery and CPC classification.
mpep-search
Expert system for searching USPTO MPEP, 35 USC statutes, 37 CFR regulations, and post-Jan 2024 updates.
refactor
Refactors code toward project-guideline compliance and testability (so /optimus:unit-test can safely increase coverage) across four analysis lenses, fanned out to parallel agents when the scope warrants it. Supports "testability" or "guidelines" focus plus flexible scoping. Read-only until the user approves the plan…