Borrowing it
Nothing to install: this file belongs to Vedant9500/OpenLMlib. 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/Vedant9500/OpenLMlib/master/.github/prompts/plan-lmlib.prompt.mdgit clone --depth 1 https://github.com/Vedant9500/OpenLMlibWrote 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/vedant9500/openlmlib/plan-lmlib)<a href="https://agentmods.dev/commands/vedant9500/openlmlib/plan-lmlib"><img src="https://agentmods.dev/badge/commands/vedant9500/openlmlib/plan-lmlib/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/commands/vedant9500/openlmlib/plan-lmlib"><img src="https://agentmods.dev/badge/commands/vedant9500/openlmlib/plan-lmlib.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 | $0.00000 | $0.08070 |
| Opus 5 | $0.00000 | $0.04035 |
| Sonnet 5 | $0.00000 | $0.01614 |
| Haiku 4.5 | $0.00000 | $0.00807 |
Grade A, and why
plan-lmlib 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 — 584 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Plan: LMlib — Cross-Project Knowledge & Research Library for LLMs
TL;DR
Build a local, filesystem-based knowledge library where AI models can store and retrieve research findings, solutions, and design decisions across projects. Start with a simple schema + SQLite + semantic search, then progressively add retrieval ranking, quality gates, and multi-index search. Integration with Glassbox as first proof-of-concept.
Research Recheck (Mar 2026)
- Storage: SQLite FTS5 provides built-in full-text search and BM25 ranking; FAISS is a similarity search library; HNSWlib provides an embedded ANN alternative; sqlite-vss shows FAISS "IDMap2" usage for stable IDs. (Sources: https://www.sqlite.org/fts5.html, https://github.com/facebookresearch/faiss, https://github.com/nmslib/hnswlib, https://github.com/asg017/sqlite-vss)
- Retrieval deltas: Anthropic Contextual Retrieval reports reduced top-20 retrieval failure rate by 35% with contextual embeddings, 49% with contextual BM25, and 67% with reranking; CRAG adds a retrieval evaluator plus decompose-then-recompose filtering; Lost in the Middle shows models use long contexts unevenly (start/end favored). (Sources: https://www.anthropic.com/news/contextual-retrieval, https://arxiv.org/abs/2401.15884, https://arxiv.org/abs/2307.03172)
- Quality gates/eval: RAGAS provides reference-free RAG evaluation including faithfulness; KILT evaluates provenance/citation. (Sources: https://arxiv.org/abs/2309.15217, https://arxiv.org/abs/2009.02252)
- Prompt injection: OWASP LLM01 details prompt injection types and mitigations (constraint, segmentation, least privilege, HITL, red-teaming); OpenAI safety best practices emphasize adversarial testing, limiting inputs/outputs, and HITL; indirect prompt injection is demonstrated in real systems. (Sources: https://genai.owasp.org/llmrisk/llm01-prompt-injection/, https://developers.openai.com/api/docs/guides/safety-best-practices, https://arxiv.org/abs/2302.12173)
Storage Architecture & Format Decision (Evidence-Based)
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 · 584 lines · 0 tokens per session scan A 407edcb3fd6b
plan-lmlib is a command published in the GitHub repository Vedant9500/OpenLMlib (2 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 8,070 tokens. 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 2026-09-11.
Other commands, from other repositories
wiki-query
Ask questions against the wiki. Synthesizes answers from wiki pages with cross-reference citations.
wiki-req
Capture and decompose a concept into atomic, traceable wiki requirements. Clarifies ambiguous requirements, splits them into atomic pieces, and persists them as wiki/requirements/ pages with status tracking.
wiki-digest
Generate a daily or weekly digest of wiki changes — new sources, pages, insights, and gaps.
wiki-status
Show wiki health overview — source count, page stats, orphan count, last activity dates.
ask
Query the MindBase wiki — cited answer using hybrid search + graph expansion. Usage: /mb:ask [-p ] [--all-projects].
search
Search decisions and checkpoints (semantic or list recent).