Borrowing it
Nothing to install: this file belongs to liuyingxuvka/FlowGuard. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/liuyingxuvka/FlowGuard/main/.agents/skills/flowguard-model-mesh/SKILL.mdgit clone --depth 1 https://github.com/liuyingxuvka/FlowGuardWrote 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/liuyingxuvka/flowguard/flowguard-model-mesh)<a href="https://agentmods.dev/skills/liuyingxuvka/flowguard/flowguard-model-mesh"><img src="https://agentmods.dev/badge/skills/liuyingxuvka/flowguard/flowguard-model-mesh/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/liuyingxuvka/flowguard/flowguard-model-mesh"><img src="https://agentmods.dev/badge/skills/liuyingxuvka/flowguard/flowguard-model-mesh.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 | $0.00033 | $0.00668 |
| Opus 5 | $0.00016 | $0.00334 |
| Sonnet 5 | $0.00007 | $0.00134 |
| Haiku 4.5 | $0.00003 | $0.00067 |
Grade A, and why
flowguard-model-mesh 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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
FlowGuard Model Mesh
Purpose
Govern ownership, evidence tiers/freshness, partitions/reattach, siblings/closure.
Entrypoint Scope
A standalone FlowGuard satellite skill; owns hierarchy, not test/code splits.
Local Material Routing
After admission read references/model_mesh_protocol.md; load references/model_mesh_partition_protocol.md, references/model_mesh_reattachment_protocol.md, or references/model_mesh_closure_protocol.md only when triggered.
Entrypoint Acceptance Map
Accept children; verify partitions/receipts; block incomplete closure.
Use When
- A relation crosses models, a model is oversized/incomplete, or a changed child/sibling/whole-flow claim needs review.
Do Not Use When
- Do not trigger on count, split tests/code, promote child-local green, or handle ordinary one-model work; use
flowguard.
Required Workflow
- Inventory hierarchy, owners, partitions, and evidence freshness. Whole-flow claims disposition all models/relations.
- Consume compact path-quality fingerprints. Reopen only changed children, ancestors, consumers, and siblings; ModelMaturation retains single-model judgment.
- Review disjointness, Child Reattachment Gate, siblings, receipts, leaves, and closure. Topology may trigger deep review, never a second path comparison.
- In blueprint scope, connect purpose, producers/consumers, realization refs to independent inventory/binding identities without copying semantics.
- Consume
flowguard.portable_refinement.v1; hand gaps/cases/receipts to alignment, test, risk, or portable owners without a second interpreter.
Hard Gates
- Model-purpose gate: freeze task-specific failure(s) and claim_boundary; bind the candidate to native good/bad-per-failure/oracle/current evidence. Reusable types are not fixed-purpose: no mode/fallback; only FlowGuard-declared checks may support completion claims. Require real FlowGuard check engine and AGENTS.md managed record; forbid a fake mini-framework.
- One logical model has one current instance; fingerprints bind model, runner, purpose, and inputs. Compile exact ownership before execution; never rerun-all.
- Multi-model replacement is one revision. Partial activation, stale head, omitted sibling, overlap, incomplete relation, stale receipt, or missing path-quality row blocks parent confidence.
- Ordinary work materializes affected topology; blueprint-wide topology needs explicit scope and never triggers repository discovery.
- Parent projections omit deep candidates/witnesses/costs.
normative_targetcannot replaceobserveduntil implementation/evidence match. - ModelMesh proves relationships, not semantics, path quality, inventory, or qualification; counts prove none.
What ships with it
9 files 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.
- .skillguard/check-manifest.json 31 KB
- .skillguard/compiled-contract.json 38 KB
- .skillguard/contract-source.json 10 KB
- agents/openai.yaml 1.5 KB
- references/model_mesh_closure_protocol.md 4.1 KB
- references/model_mesh_partition_protocol.md 2.4 KB
- references/model_mesh_protocol.md 7.9 KB
- references/model_mesh_reattachment_protocol.md 2.2 KB
- references/templates/model_mesh_prompt_template.md 3.4 KB
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 · 43 lines · 33 tokens per session scan A 9e8361dfa6f0
flowguard-model-mesh is a skill published in the GitHub repository liuyingxuvka/FlowGuard (2 stars, last pushed 14d ago), licensed MIT. It adds 33 tokens to every session and 668 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
backend-idempotency
Use this skill when the user says 'idempotency', 'idempotent', 'idempotency key', 'exactly-once', 'retry safety', 'deduplication', 'duplicate detection', 'idempotent endpoint', 'safe retry'. This skill implements idempotency keys and exactly-once semantics so retries never result in duplicate side effects. Applies to…
copilotkit-upgrade
Use when migrating a CopilotKit v1 application to v2 -- updating package imports, replacing deprecated hooks and components, switching from GraphQL runtime to AG-UI protocol runtime, and resolving breaking API changes.
notifications
Send notifications through the unified notification router.
telegram-setup
Connect a Telegram bot to the Vellum Assistant gateway with automated webhook registration and credential storage.
chat-complex-documents
Chat with and search your complex documents — ask questions, extract tables and fields, and get answers grounded in the source. Connects the hosted Unstructured Transform MCP server to parse, structure, and enrich PDFs, Word/Excel/PowerPoint, images, scanned files, emails, and 60+ other formats into clean, AI-ready…
vellum-skills-catalog
Discover bundled skills and search/install community skills from the skills.sh registry.