claudecodex

output-verification

Review structural gaps in caller-supplied claim evidence. No independent source verification; every verdict requires human review.

8 on GitHubby Vassiliy Lakhoninv1.21.2MITupdated today

Not in a marketplace yet, so it loads for one session with --plugin-dir.

Plugin output-verification in vassiliylakhonin/agenda-intelligence-md. Commands read from the repo's manifest — today.

What it installs

Skills (1)

  • evidence-output-review

    Review agent-produced claims against caller-supplied evidence before a human decides whether to relay the output.

MCP servers (1)

  • output-verification

    connects to: https://agent-output-verification-a2a.vassiliy-lakhonin.workers.dev/mcp/output-verification

View source on GitHub agent-output-verification-a2a.vassiliy-lakhonin.workers.dev

More from vassiliylakhonin/agenda-intelligence-md

All 5 plugins →

central-asia-caspian

claude

Regional specialist layer: Central Asia and Caspian cross-border risk, sanctions exposure, corridors, energy, and political economy.

5 skills

8

agenda-intelligence

claudecodex

Evidence-packet linter for claim-to-source references, quotes, lexical support, and unmatched numbers, plus compatibility agenda-analysis skills.

3 skills · 1 MCP server

8

global-think-tank-analyst

claude

Horizontal strategic-risk reasoning method: decision-useful memos on geopolitics, sanctions, trade, and regulation with explicit evidence boundaries, scenarios, and watch-next indicators.

4 skills

8

gulf-middle-east

claude

Regional specialist layer: Gulf and Middle East strategic risk, Iran/GCC dynamics, maritime and energy exposure.

4 skills

8

Maintain output-verification?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins5 plugins