Marrs

v1.6.0

Memory maintenance helper for any RAG/vector database. Includes save_memory() helper, monitor and defrag script templates, hot queue support, and configurabl...

0· 63·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description match the contents: a save_memory() helper that POSTs to /ingest plus two template agents and a config file. The declared 'requests' dependency is used. No unrelated binaries, env vars, or permissions are requested.
Instruction Scope
SKILL.md instructs the user to edit scripts/config.py and create cron jobs to run monitor/defrag. The runtime instructions and scripts only read the local config and call the configured RAG_URL; they do not access other system files or secrets. Note: once you point RAG_URL at a remote endpoint, the scripts will send the provided content there and the config prints the RAG_URL to stdout (logs).
Install Mechanism
No install spec; instruction-only with small included Python scripts. The only external package is 'requests' (installed via pip per the README), which is proportionate to making HTTP calls.
Credentials
The skill requests no environment variables, credentials, or config paths. It relies on an editable config.py for RAG_URL and collection name — this is proportionate and explicit. Be aware that placing sensitive endpoints or credentials in config.py could expose them via stdout/logs.
Persistence & Privilege
always is false and the skill does not request persistent privileges or modify other skills/configs. Cron scheduling is suggested by the docs (user action), not enforced by the skill.
Assessment
This package is a small, coherent template that will POST whatever content you give it to the RAG_URL you configure. Before installing: (1) review and edit scripts/config.py so RAG_URL points only to a trusted endpoint; (2) do not place secrets or third-party endpoints in config.py if you don't trust them (the script prints the RAG_URL to stdout/logs); (3) run the scripts in an isolated environment first and only add cron jobs after you verify behavior; (4) 'requests' is the only external dependency to install. No credentials are requested by the skill, but if you add credentials to the config or change the code to read env vars, treat them as sensitive.

Like a lobster shell, security has layers — review code before you run it.

benchmarkingvk97ckhx0h41p1knghvmt2cavdh84eachembeddingvk97ckhx0h41p1knghvmt2cavdh84eachlatestvk97811gr7zgre8g3gqx7zw3jq584k15bload-testvk97ckhx0h41p1knghvmt2cavdh84eachmarssvk97ckhx0h41p1knghvmt2cavdh84eachmemoryvk97ckhx0h41p1knghvmt2cavdh84eachperformancevk97ckhx0h41p1knghvmt2cavdh84eachscalabilityvk97ckhx0h41p1knghvmt2cavdh84eachtestingvk97ckhx0h41p1knghvmt2cavdh84each

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments