CodeBaseBuddy
Navigating through a new GitHub repository can be daunting, especially when one is unfamiliar with its structure and logic. The "CodeBase Buddy" system leverages the power of GPT's retrieval augmented generation capabilities to guide users through their first tasks in a new codebase. By analyzing the repository's structure, commit history, and documentation, it provides specific pointers on which files to modify and how to go about those changes. Users can input a task description, and the system would subsequently offer step-by-step guidance, akin to having an experienced developer by their side.
Category tags:Exact Web Answer Agent - ROCm value extraction
an agent that reads a natural-language request, navigates the given website (or a saved copy) and returns the one specific value asked for, with the page it came from. Not a summary, not a list of links. How it works. Each question runs as a fresh python3 /app/app.py process. A well-formed placeholder answer is written before any network access, so a crash still leaves parseable JSON. The agent fetches the start page (7 s timeouts), reduces HTML to readable text - spec tables kept as rows, menus and footnotes dropped - ranks passages with BM25 plus query coverage, and only then follows same-site links whose anchor text matches the request (max 8 pages, one hop, only when the start page held no answer). The answer comes from a type-aware extractor: price, percent, date, version, part code, e-mail, phone, number, name. The type is inferred from the question and respected strictly - a price question never returns a wattage; when no candidate of the expected type exists the answer is empty instead of a guess. Measurements keep their unit (192 + GB -> 192 GB), values lifted from the question are rejected, and a typed answer must clear a confidence floor. Output: /app/output/<query-id>_output.json with answer, citations, confidence, source_url. The Match 4 harness was not published, so the image follows the earlier Matches' container contract: mandated ROCm PyTorch base, code at /app/app.py, deps in /app/requirements.txt, no weights, no download at start-up, container alive throughout evaluation. The script accepts several argument spellings (--query/--question, --query-id/--id, --url, --corpus) plus positionals. Measured locally: 6-7 s per question over a page plus up to eight same-site pages; 0.3 s on a document corpus. Honest limits: extraction is lexical - no multi-hop reasoning; JavaScript pages are read as served HTML. Next iteration adds a resident small instruct model on the GPU to pick among candidate spans, with the deterministic pass as fallback.
Arber Arifi
Airlock
A containment-first security gateway that analyzes untrusted code in an isolated gVisor sandbox and returns an evidence-based security verdict.
NomadiT
SENTINEL RUN
SENTINEL RUN turns AI-generated code into real, verifiable work using disposable, resource-limited, network-isolated sandboxes powered by Vultr.
VultrZero
Blast Radius Zero — a sandboxed agent on Vultr
An agent that does real work — installs, builds, tests, scanners — inside disposable Microsandbox microVMs on Vultr, and returns every result as a hash-chained receipt. The agent touches the code, never the machine.
Hartono
Abysscan
Search a famous malware and play it as a short training game: do an ordinary job on a simulated Windows or Mac desktop, get hit by a replay of the real attack, flag the signs and get graded. Built on Vultr with sandboxed analysis and Serverless Inference.
Rokode Labs
For Innovators & Creators
Exact Web Answer Agent - ROCm value extraction
an agent that reads a natural-language request, navigates the given website (or a saved copy) and returns the one specific value asked for, with the page it came from. Not a summary, not a list of links. How it works. Each question runs as a fresh python3 /app/app.py process. A well-formed placeholder answer is written before any network access, so a crash still leaves parseable JSON. The agent fetches the start page (7 s timeouts), reduces HTML to readable text - spec tables kept as rows, menus and footnotes dropped - ranks passages with BM25 plus query coverage, and only then follows same-site links whose anchor text matches the request (max 8 pages, one hop, only when the start page held no answer). The answer comes from a type-aware extractor: price, percent, date, version, part code, e-mail, phone, number, name. The type is inferred from the question and respected strictly - a price question never returns a wattage; when no candidate of the expected type exists the answer is empty instead of a guess. Measurements keep their unit (192 + GB -> 192 GB), values lifted from the question are rejected, and a typed answer must clear a confidence floor. Output: /app/output/<query-id>_output.json with answer, citations, confidence, source_url. The Match 4 harness was not published, so the image follows the earlier Matches' container contract: mandated ROCm PyTorch base, code at /app/app.py, deps in /app/requirements.txt, no weights, no download at start-up, container alive throughout evaluation. The script accepts several argument spellings (--query/--question, --query-id/--id, --url, --corpus) plus positionals. Measured locally: 6-7 s per question over a page plus up to eight same-site pages; 0.3 s on a document corpus. Honest limits: extraction is lexical - no multi-hop reasoning; JavaScript pages are read as served HTML. Next iteration adds a resident small instruct model on the GPU to pick among candidate spans, with the deterministic pass as fallback.
Arber Arifi
Airlock
A containment-first security gateway that analyzes untrusted code in an isolated gVisor sandbox and returns an evidence-based security verdict.
NomadiT
SENTINEL RUN
SENTINEL RUN turns AI-generated code into real, verifiable work using disposable, resource-limited, network-isolated sandboxes powered by Vultr.
VultrZero
Blast Radius Zero — a sandboxed agent on Vultr
An agent that does real work — installs, builds, tests, scanners — inside disposable Microsandbox microVMs on Vultr, and returns every result as a hash-chained receipt. The agent touches the code, never the machine.
Hartono
Abysscan
Search a famous malware and play it as a short training game: do an ordinary job on a simulated Windows or Mac desktop, get hit by a replay of the real attack, flag the signs and get graded. Built on Vultr with sandboxed analysis and Serverless Inference.
Rokode Labs