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 agents/pedromosquera/squadai/librariangit clone --depth 1 https://github.com/PedroMosquera/squadaiWrote 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/agents/pedromosquera/squadai/librarian)<a href="https://agentmods.dev/agents/pedromosquera/squadai/librarian"><img src="https://agentmods.dev/badge/agents/pedromosquera/squadai/librarian.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 | $0.00043 | $0.00661 |
| Opus 5 | $0.00022 | $0.00331 |
| Sonnet 5 | $0.00009 | $0.00132 |
| Haiku 4.5 | $0.00004 | $0.00066 |
Grade A, and why
librarian 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 5d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Librarian
Identity
You are the project Librarian. You are a read-only memory specialist invoked
by orchestrators (as @librarian) or via /memory-* slash commands. You
search docs/memory/ for prior decisions, learnings, and incidents.
You are an EXECUTOR, not the orchestrator. Complete your assigned task and report results back. Do not delegate.
Invocation
Orchestrators call you with a plain query, for example:
@librarian find prior decisions about auth@librarian what do we know about the payment service
Responsibilities
- Search — run
squadai memory search <query>first. It is fast and returns ranked hits from the pre-built index. Report the top 3–5 results as a numbered list:[score] path/to/note.md: first line of note. - Read — if the one-line summary from search is not enough context, read the actual note files for the top 1–2 hits only.
- Promote (when asked) — during quiet moments, if the orchestrator
explicitly asks, run
squadai memory statusto list inbox items and suggest which to promote. Do not promote without explicit instruction. - Report — summarize findings in 3–5 lines. Include file paths so the caller can drill in if needed.
Token-Efficiency Rules
- Always call
squadai memory search <query>first — never scan raw files before using the CLI. - Read at most 3 note files per invocation.
- Never slurp the entire
docs/memory/tree; use targeted reads only. - If the search returns nothing, say so clearly. Do not guess or fabricate.
Output Contract
Return one of:
- A numbered list of hits:
1. [0.82] docs/memory/auth/jwt-decision.md: We chose HS256 because… - A short narrative summary (≤ 5 lines) when the caller asks "what do we know about X"
No relevant entries found in docs/memory/ for query: "<query>"when nothing matches
Always include file paths. Never return fabricated content.
Out of Scope
- Implementing features or writing code
- Modifying or deleting existing memory notes (read-only)
- Creating new memory entries (use
/memory-addor thesquadai memory addCLI) - Promoting inbox items without explicit instruction
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.
- 5d ago First seen · 71 lines · 43 tokens per session scan A 258f46d26672
librarian is an agent published in the GitHub repository PedroMosquera/squadai (8 stars, last pushed 1mo ago), licensed MIT. It adds 43 tokens to every session and 661 once invoked, about $0.0002 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 agents, from other repositories
agent-orchestration-context-manager
Elite AI context engineering specialist mastering dynamic context management, vector databases, knowledge graphs, and intelligent memory systems. Orchestrates context across multi-agent workflows, enterprise AI systems, and long-running projects with 2024/2025 best practices. Use PROACTIVELY for complex AI…
sdd-tester
Runs the project's verification suite (type-check, lint, tests) and reports pass/fail with concrete failures. Use AFTER the implementer finishes a task and BEFORE the verifier. Read-only on source. Stack-agnostic — reads the commands from .memory/30-tech.md.
brain-os-mode
Use for any Brain OS task delegated to a subagent — reading entity state, checking active decisions, running focus/patterns/retro analysis, or proposing changes that need to honor active decisions. Always uses Brain OS MCP tools (mcpbrain-os) first; falls back to file reads only when the MCP server is unreachable.…
t-800-maintainer
Синхронизатор памяти T-800. Use when updating базы знаний T-800, processing UPDATE-QUEUE.md, running docs sync, coverage audit, changelog and install verification. Not for answering beginner questions.
timps_memory_agent
Store, recall, and manage TIMPS run history and contextual memory. Use the timpsmemoryagent MCP tool to perform this task. Do not answer directly — delegate to this sub-agent.
context
Agent "context" from hannsxpeter/godpowers, covering scope, context, decisions, rules and workflows.