Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/lukacf/meerkatnpx agentmods add skills/lukacf/meerkat/meerkat-metaWrote 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/lukacf/meerkat/meerkat-meta)<a href="https://agentmods.dev/skills/lukacf/meerkat/meerkat-meta"><img src="https://agentmods.dev/badge/skills/lukacf/meerkat/meerkat-meta/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/lukacf/meerkat/meerkat-meta"><img src="https://agentmods.dev/badge/skills/lukacf/meerkat/meerkat-meta.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 | $0.00114 | $0.15975 |
| Opus 5.5 | $0.00046 | $0.06390 |
| Sonnet 5.5 | $0.00023 | $0.03195 |
| Haiku 4.5 | $0.00011 | $0.01597 |
Grade A, and why
meerkat-meta 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 2d 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 — 755 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Meerkat Meta
Meta is the companion composition and product layer around the Meerkat multi-agent runtime. Meerkat owns agent and mob construction, member identity and lifecycle, spawning, wiring, comms, sessions, tools, and the core agent loop. Meta composes that authority with module supervision, routing and delivery, the identity-first runtime, REST/JSON-RPC/SSE gateways, console projections, operational policy, SDKs, and examples.
Apps own desired roster and product policy. Meerkat machines admit member and session transitions. Meta translates app intent into those authorities, supervises its own operational modules, and owns product-facing projection, streaming, inspection, and console composition.
Current Baseline
Source: meta/ in the lukacf/meerkat repository, beside Core and Connect in
one Cargo workspace (releases through Meta 0.9.7 were cut from the standalone
lukacf/meerkat-meta repository). One Meta version covers every artifact:
| Artifact | Name |
|---|---|
Rust crate (library meerkat_meta) |
meerkat-meta |
Rust crate for opaque extension documents (the meerkat-meta feature extension-state) |
meta-extension-state |
Python SDK (import meerkat_meta) |
PyPI meerkat-meta |
| TypeScript SDK | npm @rkat/meta-sdk |
| Console/admin gateway | meta_gateway (release archive meta-gateway) |
| SDK stdin JSON-RPC gateway | rpc_gateway (release archive meta-rpc-gateway) |
| Operator repair of predecessor composition manifests | meta_repair (release archive meta-repair) |
crates/meta-store-conformance is a workspace-only conformance crate
(publish = false). meta_flow_editor is a development binary that CI builds;
it is not released.
The authoritative Meta release line is the version of
crates/meerkat-meta/Cargo.toml, which every crate under meta/ declares
too; the root [workspace.package].version is Core's, on its own cycle. The
0.9 line pairs with Meerkat 0.9. make verify-version-parity (from the
repository root, make -C meta verify-version-parity) checks the Meta crates'
versions, the Python and TypeScript SDK package versions, the TypeScript
lockfile's root versions, a Bazel module when meta/ has one, the primary
crates/meerkat-meta/BUILD.bazel version fields, and canonical Rust
installation snippets. Separately, run
node scripts/generate-bazel-rust-builds.mjs --check to check generated
per-crate BUILD freshness.
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.
- 2d ago Changed · +9 lines 200c5dc40068
- 3d ago Changed · +11 lines 9cbccbcb42e9
- 6d ago Changed 82ee415c4fb3
- 9d ago First seen · 735 lines · 114 tokens per session scan A 730b775b694f
meerkat-meta is a skill published in the GitHub repository lukacf/meerkat (24 stars, last pushed today), licensed Apache-2.0. It adds 114 tokens to every session and 15,975 once invoked, about $0.0005 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 2026-11-04.
meerkat-meta compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| meerkat-metalukacf/meerkat | Skill | 24 | today | Apache-2.0 | A | 114 | Claude Code |
| stuckQwenLM/qwen-code | Skill | 30k | today | Apache-2.0 | A | 57 | Claude Code |
| e2e-testingQwenLM/qwen-code | Skill | 30k | today | Apache-2.0 | A | 94 | Claude Code, Codex |
| structured-debuggingQwenLM/qwen-code | Skill | 30k | today | Apache-2.0 | A | 85 | Claude Code, Codex |
| desktop-brand-builderQwenLM/qwen-code | Skill | 30k | today | Apache-2.0 | A | 68 | Claude Code, Codex |
| batch-apiQwenLM/qwen-code | Skill | 30k | today | Apache-2.0 | A | 51 | Claude Code |
| repo-hygieneQwenLM/qwen-code | Skill | 30k | today | Apache-2.0 | A | 48 | Claude Code, Codex |