A command that records the current project work, decisions, open files, and next steps in `.claude/context.md` before a context reset.

Use it before compacting a session, then run the suggested command to continue work from the saved summary.

It gives the agent a written snapshot to use after its conversation context is shortened or cleared.

autocompact — Command for Claude Code by vanzan01

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.

Clone the repo

git clone --depth 1 https://github.com/vanzan01/claude-code-sub-agent-collective

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/commands/vanzan01/claude-code-sub-agent-collective/autocompact/github.svg)](https://agentmods.dev/commands/vanzan01/claude-code-sub-agent-collective/autocompact)

Your own site

<a href="https://agentmods.dev/commands/vanzan01/claude-code-sub-agent-collective/autocompact"><img src="https://agentmods.dev/badge/commands/vanzan01/claude-code-sub-agent-collective/autocompact/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/commands/vanzan01/claude-code-sub-agent-collective/autocompact"><img src="https://agentmods.dev/badge/commands/vanzan01/claude-code-sub-agent-collective/autocompact.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

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 283 The whole file, excluding the scripts and references it only reads on demand.

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	$0.00008	$0.00283
Opus 5.5	$0.00003	$0.00113
Sonnet 5.5	$0.00002	$0.00057
Haiku 4.5	$0.00001	$0.00028

Measured 2mo ago against content hash 9d0449bb28f4, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

autocompact 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 2mo 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.

templates/commands/autocompact.md · 41 lines

What it actually says

AutoCompact - Save Context to File

Step 1: Analyze Current State

Extract and document:

Current project/phase/task status
Todo list state (use TodoRead if available)
Recent implementations and changes
Open files and active work
Key technical decisions
Next planned steps

Step 2: Write Context File

Create/update .claude/context.md with the full context in a structured format:

# Context Summary
## Current Status
- **Project**: [project name]
- **Phase**: [current phase]
- **Task**: [active task]
## Completed Work
[List what's been done]
## Active Todos
[Current todo items with status]
## Technical Context
- **Open Files**: [files being edited]
- **Recent Changes**: [what was changed]
- **Key Decisions**: [important choices made]
## Next Steps
[What needs to be done next]

Step 3: Provide Simple Compact Command

Output:

/compact Review context and continue work @.claude/context.md

This way:

Context is automatically saved to a file
User runs one simple command that references the file
Context is preserved across compacts $ARGUMENTS

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.

2mo ago First seen · 41 lines · 8 tokens per session scan A 9d0449bb28f4

Subscribe to this mod's changes

autocompact is a command published in the GitHub repository vanzan01/claude-code-sub-agent-collective (525 stars, last pushed 6mo ago), licensed MIT. It adds 8 tokens to every session and 283 once invoked, about $0.0000 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.

Claude Code Memory and context agents ai-development claude taskmaster yolo

autocompact next to 6 commands from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
autocompactvanzan01/claude-code-sub-agent-collective	Command	525	6mo ago	MIT	A	8	Claude Code
rememberyonatangross/orchestkit	Command	299	yesterday	MIT	A	37	Claude Code, Cursor
memoryyonatangross/orchestkit	Command	299	yesterday	MIT	A	36	Claude Code, Cursor
dreamyonatangross/orchestkit	Command	299	yesterday	MIT	B	39	Claude Code, Cursor
expectyonatangross/orchestkit	Command	299	yesterday	MIT	A	51	Claude Code, Cursor
design-context-extractyonatangross/orchestkit	Command	299	yesterday	MIT	A	47	Claude Code, Cursor
design-importyonatangross/orchestkit	Command	299	yesterday	MIT	A	47	Claude Code, Cursor