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 commands/nel-neru/libraium/receptiongit clone --depth 1 https://github.com/nel-neru/LibrAIumWhat 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 | $0.00010 | $0.00799 |
| Opus 5 | $0.00005 | $0.00400 |
| Sonnet 5 | $0.00002 | $0.00160 |
| Haiku 4.5 | $0.00001 | $0.00080 |
Grade A, and why
reception scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Draft each entry's ## Reception — synthesized third-party signal (what the community reports: complaints, adopters, limitations, migration, maturity), not the owner's firsthand experience. This replaces the retired /confirm-notes: the owner is a curator, not a hands-on user of most entries, so there is no firsthand experience to confirm — honesty comes from sourcing every claim. One run covers ~3 entries.
Converse in Japanese (per CLAUDE.md); keep the entry files themselves in English.
1. Pick the next batch
Read .claude/reception-review.md. Take the first 3 unchecked (- [ ]) entries in file order. If none remain, tell the owner every entry has Reception and stop.
Show current coverage — node scripts/curation-report.mjs prints it in the reception line, or count the boxes directly. Batch is 3, not 5: each entry costs several gh calls and gh search has a ~30/min budget.
2. Gather evidence (this replaces the interview)
First export GITHUB_TOKEN=$(gh auth token) (clears the anonymous rate limit). Then run the read-only dossier for the batch:
node scripts/reception-scan.mjs --json --only <entry-id> # once per entry
It gathers — GitHub-only and writing nothing — the most-reacted issues (complaints/limitations), release cadence + open-issue count (maturity), and README adopter links. Read each dossier as your source material.
General web is an explicit opt-in. If a dossier is thin (few issues, no adopters), you MAY WebSearch for migration/positioning signal ("migrated from X to Y", "X vs Y", " in production") — but announce it first, since it reaches beyond GitHub. Never invent signal; where evidence is thin, write "limited public signal" rather than fabricate.
3. Draft in house style
Follow the entry-authoring skill's "Reception voice". For each entry, write the ## Reception section:
- 2–5 attributive bullets, third-person, every claim carrying its source — an issue #/URL, a release date, a named adopter + link. Never first person ("I", "we"), never fabricated experience, never a README/marketing paraphrase.
- Open the section with the provenance comment stamping today's date:
<!-- Third-party reception, not the owner's firsthand experience. Synthesized from public GitHub issues/releases and adopter mentions; each claim carries its source. Last gathered: YYYY-MM-DD. --> - Cross-link related library entries as Markdown links where relevant.
- Firsthand seeds (qdrant, llama_index, langgraph, modelcontextprotocol/servers, swarm, casey/just, sveltejs/kit): ADD
## Receptionabove the existing## Personal Notes— keep both. Doc-derived entries: REPLACE the old doc-derived## Personal Noteswith## 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.
- yesterday First seen · 48 lines · 10 tokens per session scan A 37fce3f6cc3c
reception is a command published in the GitHub repository nel-neru/LibrAIum (0 stars, last pushed 5d ago), licensed MIT. It adds 10 tokens to every session and 799 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
good-question
Adaptive requirement elicitation - systematically reduce uncertainty through information-maximizing questions.
patterns
Manage learned patterns: save as skills/agents, analyze, review quality, and view statistics.
learn
Learn from your work: add notes, view patterns, and build knowledge.
workflows
Manage workflows: save procedures, view/execute, and optimize quality.
active
Toggle active learning: capture prompts and file edits automatically.
help
As You: pattern learning and external memory.