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 agentmods add skills/hogu-dev/llmkb/kb-lintnpx skills add hogu-dev/llmkb --skill kb-lintgit clone --depth 1 https://github.com/hogu-dev/llmkbWrote 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/hogu-dev/llmkb/kb-lint)<a href="https://agentmods.dev/skills/hogu-dev/llmkb/kb-lint"><img src="https://agentmods.dev/badge/skills/hogu-dev/llmkb/kb-lint.svg" alt="Measured on agentmods" 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 | $0.00054 | $0.00369 |
| Opus 5 | $0.00027 | $0.00185 |
| Sonnet 5 | $0.00011 | $0.00074 |
| Haiku 4.5 | $0.00005 | $0.00037 |
Grade A, and why
kb-lint 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 3d 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.
What it actually says
kb-lint
Bring an llmkb knowledge base back to health: fix the mechanical defects the
tool finds, then the semantic ones only a reader can catch.
When to use
Periodically, or when a KB feels stale or inconsistent, or before relying on it for an important query.
Steps
- Run the tool.
llmkb lint --kb <dir>. Fix every error: broken links, pages missing from the index (llmkb reindexrepairs the index), dangling index entries, malformed frontmatter. Address warnings (orphans, log-format) where they matter. - Judgment checks the tool cannot do:
- Contradictions. Read related pages together; reconcile claims that conflict, and note the resolution.
- Stale claims. Where a newer source supersedes an older page, update the
page and its
timestamp; keep provenance insources. - Missing concept pages. If several summaries reference an idea that has no concept page, create one and link them.
- Re-run
llmkb lintand confirm it is clean; append a## [YYYY-MM-DD] lint | <scope>line towiki/log.md.
Done when
llmkb lint is error-free and the semantic pass found no unresolved
contradictions, stale claims, or missing concept pages.
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.
- 3d ago First seen · 36 lines · 54 tokens per session scan A fd7b73f54508
kb-lint is a skill published in the GitHub repository hogu-dev/llmkb (1 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 54 tokens to every session and 369 once invoked, about $0.0003 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
memory-loop-setup
Bootstrap a battle-tested memory loop (shared warm-layer BRIEF · 6-type capture log · open-loops registry · session-card hook · deadman switch · human gate) into any repo or Obsidian vault. Triggers: (1) "set up the memory loop / second brain / warm layer here", (2) "install this memory structure in another…
qector-testing-strategy
The QECTOR testing and validation strategy (manual 23). Covers the eight validation layers (syndrome faithfulness, property- based, exhaustive oracle, cross-decoder equivalence, DEM pipeline, LER parity, qLDPC correctness, memory behaviour, API stability, GPU kernels), the stale test-count policy, the cross-decoder…
qector-architecture
QECTOR architecture overview: the Rust core / Python layer FFI boundary, the module map, the threading and memory models, the layering of services, and the bit-determinism guarantee. Load for any question about the engine's internals, where a class lives, how the GIL interacts with the decode call, or why batch decode…
eng-unity-mobile-optimization
Mobile-specific Unity optimization patterns for memory, battery, thermal, and performance.
tools-unity-addressables
Unity Addressables patterns for asset loading, memory management, reference counting, and remote content delivery.
tools-unity-memorypack
MemoryPack high-performance binary serialization for Unity including custom formatters, versioning, and performance patterns.