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 Vladick-Pick/business-ontology --skill drift-flaggit clone --depth 1 https://github.com/Vladick-Pick/business-ontologyWrote 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/vladick-pick/business-ontology/drift-flag)<a href="https://agentmods.dev/skills/vladick-pick/business-ontology/drift-flag"><img src="https://agentmods.dev/badge/skills/vladick-pick/business-ontology/drift-flag/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/vladick-pick/business-ontology/drift-flag"><img src="https://agentmods.dev/badge/skills/vladick-pick/business-ontology/drift-flag.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 | $0.00035 | $0.03206 |
| Opus 5 | $0.00017 | $0.01603 |
| Sonnet 5 | $0.00007 | $0.00641 |
| Haiku 4.5 | $0.00003 | $0.00321 |
Grade A, and why
drift-flag 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 11d 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Drift flag
Purpose
A model of reality is only useful while it still matches reality, and reality moves. A definition that was accurate in March quietly rots when the team changes the refund window in June; a process card still shows three states after a fourth was added; an owner left and nobody updated the card. None of this is a bug in the ontology — it is the ontology doing its most important job: surfacing the seam between what we wrote down and what is now true. Drift is first-class here, not an embarrassment to hide.
The danger is the silent fix. The tempting move, when you spot a stale card, is to just correct it and move on. That erases the evidence that model and reality disagreed and lets the agent become the author of truth. The architecture splits the gate: the agent proposes, an authorized human decides, and the deterministic controller applies the exact revision. A drift fix therefore cannot be the agent's decision.
So this skill does something deliberately smaller than fixing. It captures the divergence as a visible, traceable candidate against the affected card id, classifies it, and routes the actual repair through propose-change into staged/ where the human can weigh it. The reasoning matters more than the rule: a flagged divergence is a question to the human about reality, and questions must stay open and attributable until the human answers — not get pre-empted by an agent's confident edit.
This is a self-initiated skill. You reach for it the moment a card stops matching what you are seeing in a source or a working session, not because someone filed a ticket asking you to.
When to use
Reach for this skill when:
- A card's
last-reviewed/next-auditcadence is overdue and a drift-sweep is checking whether the model still holds. - While mining a fresh source or working alongside the module you notice a card whose definition, states, transitions, owner, metric, or links no longer match what the source or the running system says.
- A regulation or SOP describes how something should work, and you can see practice diverging from it (or the reverse — practice has settled into a better path than the written rule).
- Two sources you trust now disagree where they used to agree, or a card cites a fact that a more authoritative source has since contradicted.
- A linked card was deprecated or its id changed, leaving a neighbor card referring to a reality that no longer exists.
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.
- 11d ago First seen · 130 lines · 35 tokens per session scan A e00f5bd5d32d
drift-flag is a skill published in the GitHub repository Vladick-Pick/business-ontology (2 stars, last pushed 1mo ago), licensed MIT. It adds 35 tokens to every session and 3,206 once invoked, about $0.0002 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
graph-mutation-plan
Cookbook for composing an applygraphmutations plan — stable entitykey patterns, the canonical label/edge vocabulary, evidence/invalidation/confidence discipline, and a worked example. Load this when building a non-trivial mutation plan.
open-ontologies
AI-native ontology engineering using 50+ MCP tools backed by an in-memory Oxigraph triple store. Build, validate, query, and govern RDF/OWL ontologies with a generate-validate-iterate loop. Use when building ontologies, knowledge graphs, RDF data, SPARQL queries, BORO/4D modeling, SHACL validation, clinical…
Rare Disease Skills Index
Skills for rare disease case support: ontology-first normalization/retrieval and the clinical genetics consult report format contract (structure + theme). Load the relevant skill file when performing the matching task.
objectstack-platform
Bootstrap, configure, extend, and operate ObjectStack runtimes. Covers project setup (defineStack, drivers, scaffolding), declaring platform capabilities (requires: — which service plugins boot), plugin and service development (PluginContext, DI, kernel hooks like kernel:ready), and operations (CLI commands…
objectstack-automation
Design ObjectStack automation — Flows (visual logic), Triggers, Approvals, state machines, and the jobs (defineJob) / webhooks (defineWebhook) stack collections. Use when the user is adding .flow.ts, wiring an event-driven rule, modelling an approval chain, or building an interactive screen flow / wizard…
objectstack-data
Design ObjectStack data schemas — objects, fields, field conditional rules, relationships, validations, indexes, lifecycle hooks, permissions, row-level security, data lifecycle retention/TTL/rotation, metadata protection locks, and external / federated datasources (defineDatasource) — and the seeds (defineSeed())…