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/nel-neru/LibrAIumnpx agentmods add commands/nel-neru/libraium/curate-reviewWrote 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/commands/nel-neru/libraium/curate-review)<a href="https://agentmods.dev/commands/nel-neru/libraium/curate-review"><img src="https://agentmods.dev/badge/commands/nel-neru/libraium/curate-review.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 5.1 | $0.00021 | $0.00867 |
| Opus 5 | $0.00010 | $0.00434 |
| Sonnet 5 | $0.00004 | $0.00173 |
| Haiku 4.5 | $0.00002 | $0.00087 |
Grade A, and why
curate-review 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 6d 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Audit the LibrAIum library (data/entries/) for curation debt. This is a review of the content, not the code.
1. Baseline
Run the structural validator first — a broken file invalidates the rest of the audit:
node scripts/validate-data.mjs --data-dir data
Report and fix any structural failures before continuing.
2. Audit checks
Run the deterministic report first — it computes checks (a), (c) and (d) in one offline pass:
node scripts/curation-report.mjs # human-readable; --json for scripting
Interpret its sections, don't just paste them:
a) Freshness — entries in the 30d+/90d+/missing buckets have unrefreshed stars/push dates. Fix via /refresh-metadata (dry-run first) or node scripts/refresh-metadata.mjs --only <entry-id> --write.
c) Tag taxonomy drift — singleton tags are rename candidates ONLY when a near-synonym candidates pair names the same concept (vector-db ~ vectordb); genuinely new singletons are fine — judge each one, don't mass-delete. Apply approved renames atomically with node scripts/rename-tag.mjs <old> <new> [--merge] (dry-run first) — never by hand-editing N files.
d) Succession — UNCOVERED stale/archived entries are shelf holes: no active same-category entry shares a tag (the suggest_alternatives rule in src-tauri/src/search.rs / alternativesFor in mcp-server/lib/suggest.js). Research a replacement or add the succession note. Also flag stale entries whose notes don't name what superseded them.
Then do the one judgment-only check the report cannot compute:
b) Placeholder Reception — a ## Reception section that is missing, empty, or contains only bare stubs (a lone - bullet, "TODO", an unsourced claim, or a bullet that merely restates the summary/README). Reception — synthesized third-party signal with a source per claim — is the library's primary content layer; flag every entry whose Reception is absent or uncited. (Entries the owner has genuinely used may also keep a firsthand ## Personal Notes; that is a bonus, never a substitute for Reception.)
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.
- 6d ago First seen · 59 lines · 21 tokens per session scan A bd3a653a74e1
curate-review is a command published in the GitHub repository nel-neru/LibrAIum (0 stars, last pushed 3d ago), licensed MIT. It adds 21 tokens to every session and 867 once invoked, about $0.0001 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 commands, from other repositories
iai-directive
Record a standing order the user typed themselves, as an explicit memory directive.
review
Comprehensive code review for blz changes.
blz
Search, retrieve, and manage documentation with BLZ.
promote
Move one or more project memories into the cross-project global store so they recall from any project.
reset_project
Wipe the per-project memory cache for the active project when it was re-cloned from the same path.
advisor
Run the advisor reflection over the active project (anchored recommendations, no remote calls).