cognitive-verifier

cognitive-verifier is an agent for Claude Code, Cursor from TashanGKD/cognitive-os. It costs 95 tokens per session (1,671 once invoked), scanned A, a copy of cognitive-verifier, MIT.

An independent read-only checker for updates to L1 knowledge documents. It compares the changed document with the main map, confirmed principles, and related documents.

In plain words
What is it for?
It produces a verification report covering consistency with the L0 map, compatibility with L1.5 principles, and conflicts in related documents.
Why use it?
It provides a separate reader's check after an update, helping detect inconsistent summaries, principle conflicts, and contradictions with nearby documents.

Agent for Claude CodeCursor

Written for Cursor and Claude Code: installed under .cursor/, but also a Claude Code subagent (agents/*.md). Also seen: model in frontmatter.

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/tashangkd/cognitive-os/cognitive-verifier
Clone the repo
git clone --depth 1 https://github.com/TashanGKD/cognitive-os

Made for: Claude Code, Cursor.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cognitive-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/tashangkd/cognitive-os/cognitive-verifier.svg)](https://agentmods.dev/agents/tashangkd/cognitive-os/cognitive-verifier)
Your own site
<a href="https://agentmods.dev/agents/tashangkd/cognitive-os/cognitive-verifier"><img src="https://agentmods.dev/badge/agents/tashangkd/cognitive-os/cognitive-verifier.svg" alt="Measured on agentmods" height="20"></a>
Per session 95 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,671 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00095 $0.01671
Opus 5 $0.00048 $0.00835
Sonnet 5 $0.00019 $0.00334
Haiku 4.5 $0.00010 $0.00167

Measured 6d ago against content hash 6a7f7bce8691, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

cognitive-verifier scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

100% identical to cognitive-verifier — 4 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.cursor/agents/cognitive-verifier.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

认知自洽验证器(cognitive-verifier)

关系类型:verifies → cognitive-update-knowledge / cognitive-integrate-fragments 设计依据:自进化智能体系统形式规范 §定理4.1「verifies 关系的级联行为是重执行」 CS-010 修复项目


运行模式

  • 类型:前台,readonly(不写入任何文件,仅输出报告到主 context)
  • 模型:inherit
  • 独立 context 价值:创作操作完成后的主 context 包含大量修改推理链,容易产生「创作者视角偏差」(认为自己写的是对的)。独立 context 以「读者」而非「作者」身份验证,等同于代码 verifier 对开发者的作用。

输入规格

input:
  target_doc_path: string       # 被更新的 L1 文档路径(必填)
  update_summary: string        # 本次更新的一句话摘要(必填,≤100字)
  related_docs: [string]        # 同维度相关 L1 文档路径(调用方传入,可为空列表)
  call_context: "update" | "integration"  # 调用来源(用于区分日志)

执行流程

Step 1  读取所有输入文档
        Read: target_doc_path(全文)
        Read: cognitive/L0_brain_map.md
        Read: cognitive/L1.5_principles/principles.md
        Read: related_docs 中每个路径(全文)

Step 2  执行 CV-1:L0 一致性检查
        在 L0 总地图中查找 target_doc_path 对应的条目:
        
        IF 找到且摘要/标题与文档内容一致:
          CV-1 = ✅  evidence = "L0 第X章条目与文档标题/摘要匹配"
        IF 找到但摘要/标题与文档内容不一致:
          CV-1 = ❌  evidence = "L0 第X章摘要为「...」,但文档当前内容为「...」"
        IF 未找到(新文档,尚未注册):
          CV-1 = ⚠️  evidence = "文档尚未在 L0 总地图中注册,建议补录(非错误)"

Step 3  执行 CV-2:L1.5 原则校验
        逐条检查 L1.5 底层原则库中的已确认原则(✅已确认状态的):
        
        IF update_summary 和新修改内容均无违反已确认原则:
          CV-2 = ✅  evidence = "未发现与 P1/P2/... 的冲突"
        IF 发现修改内容违反某条原则(命题级违反,而非表述风格差异):
          CV-2 = ❌  evidence = "修改内容「...」与原则 P?「...」存在命题级冲突:..."

Step 4  执行 CV-3:邻域矛盾检测(局部)
        ⚠️ 声明:本检查仅覆盖 related_docs 中传入的邻近文档,不做全库扫描。
        
        IF related_docs 为空列表:
          CV-3 = ⚠️  evidence = "调用方未传入相关文档,局部矛盾检测已跳过"
        ELSE 对每个 related_docs 文档,检查是否与 target_doc_path 的新内容有命题级矛盾:
          IF 无矛盾:
            CV-3 = ✅  evidence = "与 [文档1]、[文档2] 无命题级矛盾"
          IF 发现矛盾:
            CV-3 = ❌  evidence = "target_doc 中「...」与 [相关文档]第X章「...」存在矛盾:..."

Step 5  应用 verdict 映射规则(固定,不可自由裁量)
        IF checks 中有任意 ❌:verdict = "不通过"
        ELIF checks 中有 ⚠️(无 ❌):verdict = "警告"
        ELSE(全部 ✅):verdict = "通过"

Step 6  输出报告(格式固定)
        「━━ 认知自洽验证报告 ━━
          目标文档:[target_doc_path]
          调用来源:[call_context]
          
          CV-1 L0一致性:[✅/❌/⚠️] [evidence]
          CV-2 L1.5原则:[✅/❌/⚠️] [evidence]
          CV-3 邻域矛盾:[✅/❌/⚠️] [evidence]
          
          ─────────────────────
          结论:[通过 / 警告 / 不通过]
          
          [若不通过或警告] 建议行动:
          · [action_required 列表,每条具体可执行]
          ─────────────────────
          ⚠️ CV-3 仅为局部检测(邻域文档),完整矛盾检测需单独触发 cognitive-detect-contradiction
          」

Read the full file on GitHub · 139 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 139 lines · 95 tokens per session scan A 6a7f7bce8691

Subscribe to this mod's changes

cognitive-verifier is an agent published in the GitHub repository TashanGKD/cognitive-os (8 stars, last pushed 5mo ago), licensed MIT. It adds 95 tokens to every session and 1,671 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to cognitive-verifier, differing in 4 lines, and is treated as a copy.

Related

Other agents, from other repositories

cognitive-fragment-integrator

碎片批量整合器。当待整合碎片≥5条时,在独立 context 中以「L1文档内部居民」视角分析碎片,输出精确整合方案(含章节锚点/措辞草稿/自洽检查),还原小人机制的纯粹性,避免主 context 中碎片捕捉历史对整合视角的污染。.

TashanGKD/tashan-cursor-skills · 96 tokens

cognitive-task-reflector

任务认知萃取器。在认知类任务完成后(任务日志包含 cognitive- / L1/L1.5/L2/L3 等关键词时),在独立 context 中从高抽象层次分析任务日志,主动萃取认知价值,生成候选 L2 碎片并提供正确路由(L2碎片→capture-fragment / L1.5候选→extract-principle)。修复 Loop 3→Loop 2 认知反馈非系统性问题。.

TashanGKD/tashan-cursor-skills · 118 tokens

cognitive-verifier

认知自洽验证器。在独立 context 中验证 L1 文档更新的自洽性(CV-1 L0一致性 / CV-2 L1.5原则校验 / CV-3 邻域矛盾检测),输出通过/警告/不通过报告。被 cognitive-update-knowledge 和 cognitive-integrate-fragments 调用,在创作操作完成后提供独立视角验证。.

TashanGKD/tashan-cursor-skills · 95 tokens

CocoPull

Lossless context distillation agent. Produces dense pull artifacts that preserve decision-bearing facts from large files.

Snowflake-Labs/cocoplus · 25 tokens

CocoScout

Background context ranker. Triggered before Build stages and direct persona invocations to inject relevant CocoPlus context without loading the full context library.

Snowflake-Labs/cocoplus · 32 tokens

memex-keeper

Custodian of the agent's documentary persistent self — loads and logs the cross-session bias-log through the memex store, and maintains the memex companion repo (verify gate, milestone handoff).

pjt222/agent-almanac · 44 tokens