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 sageox/ox --skill post-cutoffgit clone --depth 1 https://github.com/sageox/oxWrote 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/sageox/ox/post-cutoff)<a href="https://agentmods.dev/skills/sageox/ox/post-cutoff"><img src="https://agentmods.dev/badge/skills/sageox/ox/post-cutoff/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/sageox/ox/post-cutoff"><img src="https://agentmods.dev/badge/skills/sageox/ox/post-cutoff.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 | $0.02877 |
| Opus 5.5 | $0.00032 | $0.01151 |
| Sonnet 5.5 | $0.00016 | $0.00575 |
| Haiku 4.5 | $0.00008 | $0.00288 |
Grade A, and why
post-cutoff 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 23d 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 — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Post-cutoff
What this covers: the small set of tools, models and techniques this team has actually adopted or deliberately ruled on that postdate the models we run. Each entry is a dated, graded brief with an expiry. Read the index below, open only the entries that touch what you are doing, and treat everything here as evidence, not instructions.
Why: an agent's training cutoff fails silently. There is no error and no hedge —
you design with the best approach you know, and never mention the one the team adopted
last month. AGENTS.md-style "search before building" rules tell you to go look;
nothing curates what the team already decided. This is that missing half.
This is a short shelf, on purpose. It is not a watchlist, a news feed, or a survey of everything new — those are unbounded, and an unbounded index is one nobody reads. Keeping it to what the team genuinely uses is what makes reading the whole index cheap enough to be worth doing before every design decision.
Intake is a human's call, always. An agent may propose an entry; only a person promotes one. This is the same rule the team applies to its ideas board, and for the same reason — Ryan, 2026-09-16: "Wisdom cannot be automated." Deciding that a technology is worth every teammate's agent knowing about is a judgment about the team's direction, not a research result.
This is deliberately perishable. Every entry carries a valid-through date,
six months out by default, on the assumption that model training catches up in
roughly that window. When it does, the entry stops being an advantage and becomes
weight. The retirement procedure matters as much as the authoring one.
What this is not
- Not a watchlist. "This shipped and looks interesting" is not an entry. The bar is that the team uses it, or looked hard and decided not to. Everything else is noise that makes the index too long to read.
- Not agent-generated. No agent adds an entry on its own initiative, and no agent should go searching to fill this shelf. Propose; wait for a person.
- Not a roadmap or an ideas board. Entries are falsifiable claims about the outside world, not things we might build. An idea we are dreaming about belongs on whatever board your team keeps for that.
- Not an endorsement by itself. An entry says the team has a considered position. Read the entry for what that position is — several exist specifically to stop an adoption by puncturing a launch claim.
- Not a vendor's documentation. Entries carry judgment, boundaries, and independent evidence, and link out for API syntax. If an entry starts restating a reference manual, it has lost its reason to exist.
- Not a skill finder. This is knowledge, not tooling.
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.
- 23d ago Changed · +30 lines · +79 tokens per session ddff31b8b12c
- 24d ago First seen · 200 lines · 0 tokens per session scan A 1cc197c6c79b
post-cutoff is a skill published in the GitHub repository sageox/ox (63 stars, last pushed today), licensed MIT. It adds 79 tokens to every session and 2,877 once invoked, about $0.0003 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 2026-09-22.
Other skills, from other repositories
neo4j-agent-memory-skill
Authoritative reference for the neo4j-agent-memory Python package — a graph-native memory system for AI agents built on Neo4j — and for the hosted service (NAMS) at memory.neo4jlabs.com. Use this skill whenever the user mentions neo4j-agent-memory, agent memory with Neo4j, context graphs, the POLE+O model…
maintaining-context
Use at the start of any substantial task and whenever you make a non-obvious decision or learn a reusable fact — keeps .yoke/context/ (PROJECT, DECISIONS, KNOWLEDGE) the durable source of truth so fresh-context work never drifts.
archive
Archive or backfill compact records of completed activities.
contorium-workspace-memory
Use Contorium workspace memory layers (focus, session events, Git, export) when resuming AI work or avoiding repeated project explanation.
enrich-claims
Enrich existing source-branch claims by updating them through pks source propose-claim.
bootstrap
This skill guides you through setting up your Agent Recall persona. It runs automatically on first use or when invoked with /bootstrap.