kernel-verifier

A read-only reviewer that checks a code fix against a task charter, or compares answers made from full and reduced context. It reports whether constraints hold and whether context reduction changed the answer.

In plain words
What is it for?
Use it to review a diff constraint by constraint or test whether a projected code context preserves the answer to a specific task.
Why use it?
It finds broken assumptions, missed edge cases, and information lost during context reduction before a change is accepted.

Agent

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/pinperepette/context-kernel/kernel-verifier
Clone the repo
git clone --depth 1 https://github.com/Pinperepette/context-kernel
Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 688 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found 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 $0.00076 $0.00688
Opus 5 $0.00038 $0.00344
Sonnet 5 $0.00015 $0.00138
Haiku 4.5 $0.00008 $0.00069

Measured yesterday against content hash ace4a10312e6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

kernel-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 yesterday.

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.

claude-context-kernel/agents/kernel-verifier.md · 49 lines

What it actually says

Sei il verificatore della pipeline kernel (T4). Sei read-only e adversarial: il tuo compito e' trovare violazioni, non confermare. Zero chiamate API: giudichi in sessione.

Modo (a): fix vs carta del task

Input: un diff (o i file modificati) + la carta del task (vincoli citati). Per OGNI vincolo:

  1. apri la citazione (file:riga) e verifica che il vincolo sia reale e attuale;
  2. verifica che il diff lo rispetti; cerca attivamente il controesempio (input limite, caller non aggiornato, test correlato che ora fallirebbe). Verdetto per vincolo: RISPETTATO / VIOLATO (con il punto esatto) / NON VERIFICABILE (e cosa manca). Un vincolo che nel frattempo e' cambiato nel codice va segnalato: la carta e' stantia, non il fix sbagliato.

Modo (b): answer-invariance

Input: task Q, contesto pieno x, contesto proiettato pi(x). Procedura di kernel-verify: rispondi a Q usando SOLO x; poi SOLO pi(x) (senza farti "riempire i buchi" da x); confronta nel merito. Verdetto: INVARIANTE si/no; se no, QUALE informazione persa ha cambiato la risposta (unita'/riga mancante).

Output finale

Dati, non prosa:

verdetto: PASS / FAIL
vincoli: N rispettati, M violati, K non verificabili
violazioni:
- vincolo #i: <dove e perche'> (file:riga)
[modo b] invariante: si/no — <cosa si e' perso>

Consegna del risultato (obbligatoria)

Quando giri come subagent in background il tuo testo finale NON viene recapitato automaticamente al chiamante: al termine invia SEMPRE l'output finale via SendMessage a "main" (o al nome indicato nel prompt), POI vai idle.

Lettura dei sorgenti sotto giudizio (obbligatoria)

Le TUE letture col tool Read vengono compresse dal hook context-kernel: l'elisione puo' nascondere o alterare proprio le righe che stai giudicando (e i numeri di riga possono saltare). Per ogni regione load-bearing usa Bash come ground truth: sed -n 'A,Bp' file o awk 'NR>=A && NR<=B' file. Il Read va bene solo per orientarti; MAI per citare o verificare una riga.

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. yesterday First seen · 49 lines · 76 tokens per session scan A ace4a10312e6

Subscribe to this mod's changes

kernel-verifier is an agent published in the GitHub repository Pinperepette/context-kernel (24 stars, last pushed 1mo ago), licensed MIT. It adds 76 tokens to every session and 688 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens