What is Synaptic-Tuner CLAUDE.md?

A project instruction file for Synaptic-Tuner, a system that manages shared project memory and assigns work according to the PACT framework.

It helps route tasks, load project context, manage pinned and retrieved information, and apply the correct PACT role.

It tells the AI which operating rules and startup procedure to follow before editing code or assigning work.

How ProfSynapse/Synaptic-Tuner configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to ProfSynapse/Synaptic-Tuner. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/ProfSynapse/Synaptic-Tuner/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/ProfSynapse/Synaptic-Tuner

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/profsynapse/synaptic-tuner/claude-md/github.svg)](https://agentmods.dev/instructions/profsynapse/synaptic-tuner/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/profsynapse/synaptic-tuner/claude-md"><img src="https://agentmods.dev/badge/instructions/profsynapse/synaptic-tuner/claude-md/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/profsynapse/synaptic-tuner/claude-md"><img src="https://agentmods.dev/badge/instructions/profsynapse/synaptic-tuner/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 580 This file is loaded in full into every session.

When invoked 580 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00580	$0.00580
Opus 5.5	$0.00232	$0.00232
Sonnet 5.5	$0.00116	$0.00116
Haiku 4.5	$0.00058	$0.00058

Measured 1mo ago against content hash ec3fd6e872a0, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

Synaptic-Tuner CLAUDE.md 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 1mo 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.

CLAUDE.md · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Synaptic Tuner

Synthetic dataset generation and LLM fine-tuning engine. Teacher models generate training data, which is used for SFT/KTO fine-tuning of smaller models. The engine is consumed as a git submodule by host projects; hosts own configuration, credentials, and mutable state.

How to work here

AGENTS.md carries the standing engineering discipline (fine-tuning workflow, config-first generation, HF Jobs, Modal). Read it before changing those areas.
.skills/ is the canonical skill source; .agents/skills and .claude/skills are generated mirrors. After editing a skill run python3 .skills/scripts/sync_skill_trees.py, then --check.
Load the relevant skill before starting: fine-tuning for training, cloud jobs, evaluation runs, experiments and dataset publishing; synethetic-data-generation for synthetic data; evaluation for the eval system; upload-deployment for model upload; research-reporting for research notes.
Prefer existing CLIs, checked-in scripts and skills over ad hoc scripts or manual API probing. If a capability is missing, add it as a checked-in workflow and update the skill rather than leaving a one-off.
Docs entry points: docs/common-tasks.md (how to run things), docs/troubleshooting.md (errors and recovery), docs/project-reference.md (scripts, configs, env vars, data formats).
Discover resources with the CLI rather than from memory: ./run.sh status, ./run.sh doctor, ./run.sh list <resource> (python tuner.py directly, or .\run.ps1 on Windows).

Rules

Never write outputs to /tmp. Generated files stay in the repository (docs/, Datasets/, scratch/); test outputs go under scratch/.
Stop early on errors. Watch output and kill a run as soon as it looks wrong.
No hardcoding of scenario-specific behavior in SynthChat. Tool-call formats, workspace structures and label mappings come from config under SynthChat/config/. The bundled useTools wrapper is a toy example, not the canonical format.
No backward-compatibility shims. Move code and update imports directly; no re-exports, dual signatures or deprecated wrappers.
Secrets are referenced by name only and resolved at execution time. Never put credential values in code, configs, logs, argv or commit messages.
Do not record session state, progress or status in this file. It holds only durable working rules; everything time-bound belongs in docs/.

Read the full file on GitHub · 45 lines

Changes

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.

1mo ago Changed · -221 lines · -4,458 tokens per session ec3fd6e872a0
2mo ago First seen · 266 lines · 5,038 tokens per session scan A de9b90d8efd0

Subscribe to this mod's changes

Synaptic-Tuner CLAUDE.md is an instructions file published in the GitHub repository ProfSynapse/Synaptic-Tuner (28 stars, last pushed 3d ago), licensed MIT. It adds 580 tokens to every session, about $0.0023 per session on Opus 5.5. 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 2026-08-30.

Memory and context

Synaptic-Tuner CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
Synaptic-Tuner CLAUDE.mdProfSynapse/Synaptic-Tuner	Instructions file	28	3d ago	MIT	A	580	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

Synaptic-Tuner: Instructions file for Claude Code