Search results

security: LLM meta-analysis can suppress high-severity static findings via prompt injection

Status: Open.

#59 In NVIDIA/SkillSpector;

· AbhiramDwivedi opened on Jun 14, 2026

ci: add a public GitHub Actions workflow (lint, unit tests, DCO)

Status: Open.

#58 In NVIDIA/SkillSpector;

· AbhiramDwivedi opened on Jun 14, 2026

feature: support local agent CLIs (claude/codex) as an LLM provider without an API key

Status: Open.

#57 In NVIDIA/SkillSpector;

· AbhiramDwivedi opened on Jun 14, 2026

docs: document the machine-readable integration contract and trust model

Status: Open.

#56 In NVIDIA/SkillSpector;

· AbhiramDwivedi opened on Jun 14, 2026

docs: DEVELOPMENT.md lists implemented analyzers as stubs; dangling SADD references

Status: Open.

#55 In NVIDIA/SkillSpector;

· AbhiramDwivedi opened on Jun 14, 2026

[phantomstars] Fake engagement detected on this repository

Status: Open.

#47 In NVIDIA/SkillSpector;

· tg12 opened on Jun 14, 2026

SC4 (OSV known vulnerable dependency) reports all historical CVEs when requirements.txt uses >= operator #42

Status: Open.

#43 In NVIDIA/SkillSpector;

· woxinsj opened on Jun 13, 2026

SC4 (OSV known vulnerable dependency) reports all historical CVEs when requirements.txt uses >= operator

Status: Open.

#42 In NVIDIA/SkillSpector;

· woxinsj opened on Jun 13, 2026

Bidirectional control characters (Trojan Source, CVE-2021-42574) not detected in file contents

Status: Open.

#39 In NVIDIA/SkillSpector;

· asadbekXodjayev opened on Jun 13, 2026

OpenAI fallback uses active provider default model

Status: Open.

#38 In NVIDIA/SkillSpector;

· keshprad opened on Jun 13, 2026

Creates false positives on pure text-based content

Status: Open.

#37 In NVIDIA/SkillSpector;

· francoisdp opened on Jun 13, 2026

Expose SkillSpector as an MCP server so agents can scan skills as a tool

Status: Open.

#35 In NVIDIA/SkillSpector;

· CharmingGroot opened on Jun 13, 2026