save-chat

A command for saving a Claude Code conversation as a Markdown, text, or HTML document.

In plain words
What is it for?
It is for saving the whole chat or selected turns, appending to an existing file, and optionally using the original session data for a more complete export.
Why use it?
It gives you a local record of the discussion without needing to copy the messages by hand.

Command

Install

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.

agentmods
npx agentmods add commands/dropfan/claude-code-plugins/save-chat
Clone the repo
git clone --depth 1 https://github.com/DropFan/claude-code-plugins
Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,343 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00008 $0.02343
Opus 5 $0.00004 $0.01171
Sonnet 5 $0.00002 $0.00469
Haiku 4.5 $0.00001 $0.00234

Measured yesterday against content hash 415e0ef31b6a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

save-chat 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.

plugins/chat-saver/commands/save-chat.md · 218 lines

How it starts

The opening of the file, as written. The whole thing — 218 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Context

  • Working directory: !pwd
  • Current date: !date '+%Y-%m-%d %H:%M'
  • Plugin root: ${CLAUDE_PLUGIN_ROOT}

Task

Save the current conversation to a document file. Parse user arguments and guide the export process.

User arguments: $ARGUMENTS

Step 0: Load Settings

Load settings from .claude/chat-saver.local.md following ${CLAUDE_PLUGIN_ROOT}/skills/conversation-export/references/settings-loading.md. Extract: default_format, default_scope, save_dir, custom_header, custom_footer.

Step 1: Parse Arguments

Parse the optional arguments from $ARGUMENTS:

  • formatmd (default), txt, or html
  • scopefull (default) or summary
  • --append — Append to an existing file instead of creating a new one
  • --last N — Only save the last N conversation turns (each User + Assistant pair = 1 turn)
  • --from "keyword" — Save from the first turn containing the keyword to the end
  • --raw — Use JSONL session file as data source instead of model memory (ensures 100% complete export, recommended for long conversations)

If $ARGUMENTS is empty, use defaults (md, full, no append, no raw).

If arguments are ambiguous or invalid, use AskUserQuestion to let the user choose:

For format:

  • md — Markdown format, preserves code blocks and formatting
  • txt — Plain text, clean and portable
  • html — HTML with styled layout, viewable in browser

For scope:

  • full — Complete conversation history
  • summary — Key decisions, code changes, and insights

Step 1.5: Filter Conversation Range

If --last or --from is specified, filter the conversation before processing:

--last N:

  • Count conversation turns (each User + Assistant pair = 1 turn)
  • Keep only the last N turns
  • If N exceeds total turns, keep the entire conversation (no error)

--from "keyword":

  • Scan conversation turns from the beginning
  • Find the first turn where either the user message or assistant response contains the keyword (case-insensitive)
  • Keep that turn and all subsequent turns
  • If no turn matches, inform the user and ask whether to save the full conversation instead

Read the full file on GitHub · 218 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.

  1. yesterday First seen · 218 lines · 8 tokens per session scan A 415e0ef31b6a

Subscribe to this mod's changes

save-chat is a command published in the GitHub repository DropFan/claude-code-plugins (7 stars, last pushed 26d ago), licensed MIT. It adds 8 tokens to every session and 2,343 once invoked, about $0.0000 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.