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 arunveersingh/ai --skill day-07-as-of-or-stopgit clone --depth 1 https://github.com/arunveersingh/aiWrote 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/arunveersingh/ai/day-07-as-of-or-stop)<a href="https://agentmods.dev/skills/arunveersingh/ai/day-07-as-of-or-stop"><img src="https://agentmods.dev/badge/skills/arunveersingh/ai/day-07-as-of-or-stop/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/arunveersingh/ai/day-07-as-of-or-stop"><img src="https://agentmods.dev/badge/skills/arunveersingh/ai/day-07-as-of-or-stop.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 | $0.00106 | $0.00885 |
| Opus 5.5 | $0.00042 | $0.00354 |
| Sonnet 5.5 | $0.00021 | $0.00177 |
| Haiku 4.5 | $0.00011 | $0.00089 |
Grade A, and why
as-of-or-stop 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 16d 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
As of, or stop
You ask your agent to pin the latest requests: lock it to one exact version so every install gets the same code. It writes requests==2.31.0 (latest stable). That was true in 2023. On 9 Oct 2026 the newest is 2.34.2, and 2.31.0 has three known advisories. Nothing errors and the tests pass.
Known then is not true now. The fix has two parts. Ask: anything that can change needs a date and a live source, memory alone means stop, and versions come from the package manager. Verify: flag undated claims, and let an audit against today's advisory data decide "safe."
Paste Part A into the agent's standing instructions or before the task. Do Part B yourself. Fill in the brackets.
Part A — Ask (paste and adapt)
Today is [YYYY-MM-DD]. Installed versions: [paste lockfile or
"pip freeze"].
1. For anything that can change over time (versions, prices,
limits, API behavior, people in roles), write:
as of <date>, source: <tool output or URL>
2. If your only source is training data, write FROM MEMORY.
Don't act on it.
3. If you have a lookup tool, run it and quote the output.
If not, write STOPPED: no live source. You may propose
one lookup.
4. Add dependencies with the package manager. Report the version
it installed. Never type a version number from memory.
Adapting it: for npm, Maven, or Go, swap in that package manager and lockfile. For non-code facts (pricing, rate limits, org charts), the source is a URL or document with its own date; "as of" is the date of that source, not today.
Part B — Verify (you do this)
- Flag undated claims. Reply saved as
reply.txt:! grep -nEi '\b(latest|current|newest|up to date)\b' reply.txt | grep -viE 'as of [0-9]{4}|FROM MEMORY'It prints any "latest" line with no date or label and fails if there is one. - Audit the lockfile. Pin everything, including sub-dependencies (
pip freeze > requirements.txtor pip-compile), thenpip-audit -r requirements.txt. Exit 0 = no known advisory today. Exit 1 = advisory found, or the database couldn't be reached (fails closed). No saved state, so it works the same on a fresh machine. - Let CI decide, on PRs and nightly. Require the audit on PRs and run it on a schedule, so advisories published after merge still surface. Multi-module or mixed stacks:
osv-scanner scan source -r .audits every standard lockfile under the repo in one run; exit 128 if it finds none. Ignore an advisory only with its ID and reason in a reviewed commit. Don't gate on "newest version": it flakes every release. - Inject the date in code. If you build the agent, put today's date and the current lockfile in the system prompt on every run, and give it a version-lookup tool.
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.
- 16d ago First seen · 45 lines · 106 tokens per session scan A b9f81badd52a
as-of-or-stop is a skill published in the GitHub repository arunveersingh/ai (3 stars, last pushed 3d ago), licensed Apache-2.0. It adds 106 tokens to every session and 885 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-10-09.
Other skills, from other repositories
chinese-git-workflow
A reference for configuring Git with Chinese code-hosting services such as Gitee, Coding.net, GitLab China, and CNB, including SSH, HTTPS, credentials, CI, and repository mirroring.
atmos-profiles
Atmos profiles: profile directories, --profile and ATMOSPROFILE activation, profile merge behavior, environment switching, and routing profile-specific auth/toolchain/config overrides.
webhook-management
Configure and validate CCAM webhook targets across supported chat, incident, automation, and generic providers. Use when listing provider requirements, creating or updating a target, scoping it to alert rules, sending a test notification, reviewing delivery history, or deleting a target.
monorepo-management
Master monorepo management with Turborepo, Nx, and pnpm workspaces to build efficient, scalable multi-package repositories with optimized builds and dependency management. Use when setting up monorepos, optimizing builds, or managing shared dependencies.
managing-github-actions-secrets
Creates and updates GitHub Actions secrets for PostHog workflows. Use when adding a new CI secret, rotating an existing secret, wiring a workflow to an API token, package registry credential, deploy key, or any value referenced via ${{ secrets. }} in .github/workflows/.
deployment-engineer
Expert deployment engineer specializing in modern CI/CD pipelines, GitOps workflows, and advanced deployment automation. Masters GitHub Actions, ArgoCD/Flux, progressive delivery, container security, and platform engineering. Handles zero-downtime deployments, security scanning, and developer experience optimization.…