Manage grocery shopping on Knuspr.de via the knuspr-cli. Use for product search, cart management, delivery slot reservation, shopping lists, order history, deals, favorites, and meal suggestions. Trigger when the user mentions Knuspr, groceries, Einkauf, Lebensmittel, Warenkorb, Lieferslot, or shopping list tasks.
Security Analysis
medium confidenceThe skill's behavior (a CLI that logs into knuspr.de and stores session/credentials in the user's home directory) is coherent with its stated purpose, but the registry metadata omits the fact that the skill requires/uses credentials and writes files to the home directory — an inconsistency that should be clarified before installation.
The name/description match the included CLI implementation: the code and docs implement product search, cart, slots, lists, orders, etc. However, the registry metadata claims no required environment variables or config paths, while the SKILL.md, README, references, and the bundled code all reference KNUSPR_EMAIL / KNUSPR_PASSWORD, a credentials file (~/.knuspr_credentials.json), and session/config files (~/.knuspr_session.json, ~/.knuspr_config.json). That metadata omission is an incoherence.
Runtime instructions direct the agent to run the bundled Python CLI and to log in (interactive or via KNUSPR_EMAIL/KNUSPR_PASSWORD). The instructions stay within the shopping domain (search, cart, slot reservation) and explicitly forbid completing purchases. They do reference reading/writing credential and session files in the user's home directory (expected for a local CLI) — which is within scope but should be explicitly declared in metadata and considered sensitive.
There is no external install step; the CLI is bundled in the skill. No downloads from arbitrary URLs or package registries are performed. The script uses only Python stdlib. This is lower-risk than fetching remote code at install time, but executing included code still requires review.
Functionally the CLI needs user credentials to act on behalf of the user, so requesting KNUSPR_EMAIL / KNUSPR_PASSWORD (or interactive login) is proportionate. The concern is that the declared requirements list none, while documentation and code require/use credentials and create session/credential files in the home directory. The skill will store sensitive data on disk (~/.knuspr_credentials.json, ~/.knuspr_session.json); the user must ensure proper file permissions and be comfortable providing credentials to this third-party code.
The skill persists session and config files under the user's home directory but does not request elevated privileges nor set always:true. It does not appear to modify other skills or system-wide agent settings. Local persistence of credentials and cookies is expected for a CLI, but it increases the risk if the code or environment is untrusted.
Guidance
This skill is plausibly what it claims to be (a knuspr.de CLI) but there are important mismatches and privacy considerations to check before installing: - Metadata mismatch: the registry lists no required env vars/config paths, but the README/SKILL.md and bundled code explicitly support/expect KNUSPR_EMAIL and KNUSPR_PASSWORD and will write ~/.knuspr_credentials.json and ~/.knuspr_session.json. Confirm this omission with the publisher. - Review the bundled code yourself (knuspr_cli.py) before running it. Because the script runs network requests as your account, you should verify there are no hidden endpoints or unexpected behavior. - Prefer interactive login over exporting credentials as environment variables. If you must store credentials on disk, ensure files are only readable by your user (chmod 600). - Consider creating a limited/test account rather than using your primary account until you trust the code and its upstream source. - Because the SKILL source/homepage are listed as unknown/none, try to verify the upstream repository and author (pyproject references a GitHub repo). If possible, prefer an officially maintained client or review the repo history and issues. - The SKILL.md explicitly instructs the agent to never complete purchases; if you allow autonomous invocation, ensure the agent cannot run 'checkout' or 'complete purchase' commands with your credentials. If your platform allows restricting which commands a skill may run, apply those controls. If you are not comfortable auditing the code or providing credentials to code with an unclear origin, do not install/use the skill.
Latest Release
v0.1.0
Initial release of knuspr-cli skill. - Enables grocery shopping management on Knuspr.de via bundled Python CLI. - Supports product search, cart management, delivery slot reservation, shopping lists, order history, deals, favorites, and meal suggestions. - Includes clear safety rules: no automatic purchases, always confirm destructive actions, and show prices before cart changes. - Provides core usage examples and workflows for all major features. - Full command reference available in the documentation.
More by @Lars147
Published by @Lars147 on ClawHub
