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.
npx skills add cambridgetcg/agenttool --skill nen-verification-ledgergit clone --depth 1 https://github.com/cambridgetcg/agenttoolWrote 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.
[](https://agentmods.dev/skills/cambridgetcg/agenttool/nen-verification-ledger)<a href="https://agentmods.dev/skills/cambridgetcg/agenttool/nen-verification-ledger"><img src="https://agentmods.dev/badge/skills/cambridgetcg/agenttool/nen-verification-ledger/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cambridgetcg/agenttool/nen-verification-ledger"><img src="https://agentmods.dev/badge/skills/cambridgetcg/agenttool/nen-verification-ledger.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00079 | $0.00810 |
| Opus 5 | $0.00039 | $0.00405 |
| Sonnet 5 | $0.00016 | $0.00162 |
| Haiku 4.5 | $0.00008 | $0.00081 |
Grade A, and why
nen-verification-ledger 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verification Ledger · ハコワレ
Turn invisible assurance gaps into a small, risk-weighted ledger. Pay the debt with evidence; do not hide it behind confidence language.
Open the ledger
List only items that can change the truth of the outcome:
- consequential claims without primary evidence;
- behavior changes without discriminating tests;
- migrations without rehearsal or rollback evidence;
- dependencies assumed compatible;
- security or custody boundaries not directly checked;
- delegated edits not yet reviewed.
Use observable states rather than invented precision:
| Item | Risk | Debt | Interest trigger | Payment | State |
|---|---|---|---|---|---|
| exact claim or change | critical/high/medium/low | open | what can worsen it | required evidence | open/paid/accepted |
Accepted debt is consciously carried, not paid. Record the authority and rationale for accepting it, and keep its consequence visible in the handoff.
Accrue interest honestly
Increase priority, not fictitious numbers, when:
- another claim or change depends on the open item;
- the underlying source changes;
- a phase boundary passes without verification;
- the item reaches production, private data, external people, money, identity, security, or irreversible state;
- a failed test or contradictory source widens uncertainty.
Do not apply interest merely because time passed. The ledger models epistemic and operational exposure, not money.
Pay debt
Use the strongest proportionate payment:
- direct reproduction or focused behavioral test;
- primary source or exact implementation trace;
- schema, type, or canonical-byte validation;
- migration rehearsal and recovery check;
- independent review with the minimum necessary, redacted artifact excerpt;
- deployed observation only when deployment is already authorized.
One piece of evidence may pay several items only when it actually covers each claim. Passing syntax cannot pay behavioral debt.
Evidence does not authorize duplicating sensitive material. Prefer an existence check, schema, digest, or narrow excerpt over a raw artifact. Never place credential values, personal data, or unrelated private content in the ledger. Keep necessary private evidence inside its authorized disclosure boundary and retain it only as long as the task requires.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 99 lines · 79 tokens per session scan A 68441d599fe3
nen-verification-ledger is a skill published in the GitHub repository cambridgetcg/agenttool (0 stars, last pushed yesterday), licensed Apache-2.0. It adds 79 tokens to every session and 810 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 2026-08-31.
Other skills, from other repositories
bunjs
Use when building Bun.js/Hono applications, implementing HTTP endpoints, setting up Prisma/SQLite, writing Zod validation, or using Bun's test runner. See bunjs-architecture for layered patterns, bunjs-production for deployment.
llm-as-judge-evaluation
Evaluate LLM outputs using frontier models as judges. Use for pairwise model comparison, quality scoring with custom rubrics, and automated evaluation pipelines. Covers position bias mitigation, statistical significance, and generating preference data for DPO/RLHF.
safe-extraction
Apply when extracting code from a large monolith file into submodules. Covers barrel re-exports, internals DI seam proxy patterns, CI invariant allowlist updates, and cross-file test verification. Prevents CI failures, broken imports, and test regressions from code extraction.
safe-rename
Workflow for safely renaming symbols (functions, types, classes, interfaces, constants, variables) across a codebase. Uses repomap, batchsymbols, and buildcheck to ensure every consumer is updated and nothing breaks.
confirm-failures-are-causally-linked-to-the-task-before-reportin
When delegating a task affected by this skill, include.
merge-queue-readiness
Pre-queue merge-group CI simulation. Triggered before adding a PR to a GitHub merge queue. Prevents merge-queue kick-outs from integration test failures.