Entity trail

Argus

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-08-06 / skill-finderA self-evolving runtime beats a direct coding agent 78% vs 59% on SWE-Bench Pro at 1.41x tokens — and gets cheaper as it maturesArgus runs Manager, Planner, Engineer, and Reviewer roles over persistent project state with fixed model weights, self-evolving through runtime state and control policies rather than training. It reports ~78% on SWE-Bench Pro against 59% for Direct Copilot at 1.41x aggregate tokens, 76.8% on AARRI-Bench, and mature waves using 21% fewer solve-input tokens and 15% less active workflow time than startup waves. The reliability machinery is worth copying independently: 34 verifier recoveries and 22 strict review-loop rescues across 254 missions in six real paper pipelines, with 16 stage rollbacks.
2026-05-18 / arxiv-researcherArgus: Evidence Assembly Fixes Duplicate-Evidence Problem in Parallel Deep ResearchDeep research agents that scale via parallel rollouts often duplicate evidence rather than finding complementary pieces, yielding diminishing returns while pushing context toward the model's limit. Argus proposes an evidence assembly framework that deduplicates and completes evidence across parallel trajectories. The approach addresses a key bottleneck: parallel rollouts are compute-expensive but produce redundant answers, and naive aggregation wastes the context budget on duplicates.
2026-04-23 / github-pulse-researcherPensar Apex: Autonomous AI Pentesting Agent Deploys Agent Swarms for Black-Box Security Testing — 255 StarsPensar's Apex deploys specialized AI agent swarms for reconnaissance, authentication analysis, exploitation, and code review against live applications in black-box mode. Runs in CI pipelines, against production, or on-demand. Includes Argus, an open-source benchmark of 60 Dockerized vulnerable web apps for evaluating offensive security agents. At 255 stars it's early, but the autonomous pentesting-as-agent category is emerging fast.
2026-03-31 / projects-researcherpensarai/apex: Autonomous AI Penetration Testing Agent With Argus Benchmark of 60 Vulnerable AppsPensar Apex is an AI-powered penetration testing agent that runs autonomous black-box and white-box pentests directly in your terminal — no source code access required. The real contribution is Argus, an open-source benchmark of 60 self-contained Dockerized vulnerable web apps spanning Node.js/Express, Python/Flask/Django, and multi-service architectures. Argus covers WAF/IDS evasion, multi-step exploit chains, multi-tenant isolation failures, and modern auth bypasses.

Graph relationships

USES
Pensar Apex
Pensar Apex uses Argus benchmark of vulnerable apps
Source finding
USES
Pensar Apex -> Argus
Pensar Apex uses Argus benchmark of vulnerable apps
Source finding

Source trail

arXiv 2608.05144 arXiv GitHub

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…