Skip to content

opensite-ai/opensite-skills / deep-research-agent-architecture

deep-research-agent-architecture

Unverified

Design Forge's Phase 4B deep-research subsystem so every claim in a generated report traces to a fetched source. Load when building or reviewing a multi-source research agent, planner/executor separation, parallel subagent fan-out, source adapters over src/scraping/ tiers, citation verification, claim-level entailment or faithfulness checks, hard-fail on unverifiable citations, report artifacts in Tigris, context-budget management for long syntheses, per-request cost caps for research aliases, or any request shaped as 'add deep research' / 'generate a cited report' / 'why are the citations wrong'.

Security report

Unverified

Static analysis of one exact version. It reduces risk and shows evidence; it does not certify that a skill is safe.

Commit
89248bf7c733a83a929b63ca2d68e7e41c9db25b
Content hash
778a4721a0413efed7735d0e549c83cf93facd3d88e5e24c6d782139ddeb874d
Scanned
2026-11-17
Scanner version
skillxone-scan-1.0.0
Policy version
0.1-draft
Decision
2 open medium findings to review.
Verification expires
Not applicable

Findings (2 medium)

  • The cloud instance metadata service is contacted. On a cloud machine it hands out credentials. Severity lowered from high because it appears in prose rather than code.

    Have the planner emit `http://169.254.169.254/`, a Fly 6PN `.internal` host, and a

    Suggested fix: Remove access to the metadata service.

    Report as false positive

  • mediumNET-CLOUD-METADATASKILL.md:364

    The cloud instance metadata service is contacted. On a cloud machine it hands out credentials. Severity lowered from high because it appears in prose rather than code.

    7. **SSRF and robots test.** Have the planner emit `http://169.254.169.254/`, a `.internal`

    Suggested fix: Remove access to the metadata service.

    Report as false positive

Checks run

  • prompt-injection
  • dangerous-commands
  • credentials
  • network
  • file-access
  • integrity

Checks skipped

  • dependencies: No dependency manifest in the skill directory.
  • gitleaks: gitleaks is not installed on this host.
brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 322kUpdated Claude CodeCodexVerifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 198kUpdated Claude CodeCodexVerifiedInstall