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 instructions/youngde811/lisa/claude-mdgit clone --depth 1 https://github.com/youngde811/LisaWrote 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/instructions/youngde811/lisa/claude-md)<a href="https://agentmods.dev/instructions/youngde811/lisa/claude-md"><img src="https://agentmods.dev/badge/instructions/youngde811/lisa/claude-md.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 | $0.02440 | $0.02440 |
| Opus 5 | $0.01220 | $0.01220 |
| Sonnet 5 | $0.00488 | $0.00488 |
| Haiku 4.5 | $0.00244 | $0.00244 |
Grade A, and why
Lisa CLAUDE.md scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
test-culture-1.sh — End-to-end bridge test (curl-based) How it starts
The opening of the file, as written. The whole thing — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lisa — Lisp-based Intelligent Software Agents
Forward-chaining expert system shell in Common Lisp (Rete algorithm, CLOS/MOP, certainty factors). Integrated with Claude via tool-use for natural-language medical diagnosis (MYCIN rulebase).
Build & Load
Requires SBCL with Quicklisp. From the SBCL REPL at project root:
;; Load Lisa core
(load "lisa.asd")
(asdf:load-system :lisa)
;; Load the MYCIN rules (classes + rules, no assertions)
(in-package :lisa-user)
(load "examples/mycin.lisp")
;; Load and start the LLM bridge
(asdf:load-system :lisa-bridge)
(lisa-bridge:start) ; starts on port 8090
To stop: (lisa-bridge:stop)
Choosing a belief system
The bridge honors LISA_BELIEF_SYSTEM at startup. Dempster-Shafer is the
default — it exposes ignorance intervals {bel, pl, ignorance} that the
LLM can narrate meaningfully. Override with the env var:
LISA_BELIEF_SYSTEM=cf sbcl ... # certainty factors (Shortliffe-Buchanan)
LISA_BELIEF_SYSTEM=ds sbcl ... # Dempster-Shafer (default)
Per-session overrides ride on POST /reset with body {"belief_system": "cf" | "ds"}. /conclusions echoes the active system in its response
and emits {bel, pl, ignorance} payloads under DS.
Project Structure
lisa.asd — Core system definition (depends on log4cl)
lisa-bridge.asd — Bridge system (depends on lisa, hunchentoot, jzon, bordeaux-threads)
src/
core/ — Rete engine, rules, facts, conflict resolution
belief-systems/ — Pluggable belief-system protocol
protocol.lisp — Generic function surface + dispatcher + use-system
certainty-factors/— Shortliffe-Buchanan CF implementation
dempster-shafer/ — Simplified [Bel, Pl] interval implementation
rete/reference/ — Rete network nodes and compiler
llm/bridge/ — HTTP bridge for LLM integration
package.lisp — :lisa-bridge package
session.lisp — Entity registry, session reset
server.lisp — Hunchentoot start/stop; LISA_BELIEF_SYSTEM env var
handlers.lisp — REST endpoints (belief-system-aware)
llm/claude/ — Claude tool-use integration
driver.py — Python client: tool-call loop; transcript capture
tools.json — Tool schemas (assert_fact, run_inference, get_conclusions, etc.)
system-prompt.md — Clinical diagnostic system prompt (18 rules, CF/DS output)
examples/
mycin.lisp — MYCIN rules (18 rules incl. 3 disconfirming; culture-1, culture-1a, culture-2, culture-3)
bin/
test-culture-1.sh — End-to-end bridge test (curl-based)
run-mycin.sh — Same as test-culture-1.sh (legacy name)
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 · 202 lines · 2,440 tokens per session scan A 7ee9978d37dd
Lisa CLAUDE.md is an instructions file published in the GitHub repository youngde811/Lisa (141 stars, last pushed 1mo ago), licensed MIT. It adds 2,440 tokens to every session, about $0.0122 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
autolith AGENTS.md
AGENTS.md instructions for lambda-symbolics/autolith, covering repository guidelines, purpose and sources of truth, upstream references, architectural guardrails and package policy.
sema AGENTS.md
AGENTS.md instructions for sema-lisp/sema, covering agents.md — sema (lisp with llm primitives, in rust), build & test, architecture (cargo workspace), key design patterns and writing style — plain technical english (mandatory).
elle AGENTS.md
AGENTS.md instructions for elle-lisp/elle, covering elle, architecture, key modules, the value type and signals and capabilities.
sema CLAUDE.md
Claude Code instructions for sema-lisp/sema, a project described as: A Lisp with first-class LLM primitives, implemented in Rust.
convex AGENTS.md
AGENTS.md instructions for Convex-Dev/convex, covering agent guidelines for convex, project overview, build system, agent tooling and specifications.
convex CLAUDE.md
Claude Code instructions for Convex-Dev/convex, a project described as: Convex Main Repository - Decentralised platform for the Internet of Value.