signal-compress

signal-compress is a skill for Claude Code from mattbaconz/signal. It costs 76 tokens per session (698 once invoked), scanned A, original, MIT.

A skill that shortens memory files, prompts, rules, handoff notes, and project documentation while preserving technical details. It keeps items such as code, paths, errors, versions, and identifiers intact.

In plain words
What is it for?
Shrinking files such as `CLAUDE.md`, `GEMINI.md`, and `AGENTS.md`, preparing handoff notes, and reducing prompt size before an agent reads the material.
Why use it?
Repeated project instructions consume space in an agent’s input and can make important details harder to find. Compression reduces the prose while retaining the technical substance.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions CLAUDE.md; mentions AGENTS.md; mentions Gemini CLI.

Part of the signal plugin — 8 skills shipped together

Good fit Shrinking files such as CLAUDE.md, GEMINI.md, and AGENTS.md, preparing handoff notes, and reducing prompt size before an agent reads the material.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/mattbaconz/signal/signal-compress
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.

Any agent
npx skills add mattbaconz/signal --skill signal-compress
Clone the repo
git clone --depth 1 https://github.com/mattbaconz/signal

Made for: Claude Code.

Or install signal, the plugin that ships this one along with the rest of its 8 skills.

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.

agentmods badge for signal-compress

README.md
[![agentmods](https://agentmods.dev/badge/skills/mattbaconz/signal/signal-compress/github.svg)](https://agentmods.dev/skills/mattbaconz/signal/signal-compress)
Your own site
<a href="https://agentmods.dev/skills/mattbaconz/signal/signal-compress"><img src="https://agentmods.dev/badge/skills/mattbaconz/signal/signal-compress/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.

agentmods 80×15 button for signal-compress

Your own site · 80×15
<a href="https://agentmods.dev/skills/mattbaconz/signal/signal-compress"><img src="https://agentmods.dev/badge/skills/mattbaconz/signal/signal-compress.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 76 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 698 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00076 $0.00698
Opus 5 $0.00038 $0.00349
Sonnet 5 $0.00015 $0.00140
Haiku 4.5 $0.00008 $0.00070

Measured 11d ago against content hash 88a2c483ad01, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

signal-compress 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 11d 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-signal/skills/signal-compress/SKILL.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

⚡ signal-compress — Input Compression

Reduce prompt tokens before the agent reads the same prose every turn. Preserve technical substance; compress prose only.


Invocation Triggers

Activate when user says any of:

  • /signal-compress
  • "compress this memory", "shrink this prompt", "compress project rules"
  • "compress CLAUDE.md", "compress GEMINI.md", "compress AGENTS.md"
  • "reduce input tokens", "make this rules file shorter"

Targets can be memory files, project rules, prompts, handoff notes, README-like prose, or pasted text.


Behavior

  1. Back up first when a file path is provided:

    powershell -NoProfile -ExecutionPolicy Bypass -File .\scripts\signal-compress.ps1 -Path .\GEMINI.md
    

    The helper only creates <name>.original.<ext> and prints the prompt/review steps.

  2. Compress prose only using templates/signal-compress-prompt.md.

  3. Protect exact tokens:

    • Fenced code blocks and shell commands
    • Inline code, file paths, line numbers, URLs
    • Quoted errors, stack traces, logs
    • Version numbers, dates, identifiers, flags, env vars
    • Technical terms and proper nouns
  4. Diff review against the original. Reject the rewrite if protected tokens changed, facts disappeared, or instructions became ambiguous.

  5. Report measured savings with this shape:

    signal-compress|target=GEMINI.md|chars 2400→980|est_tok 600→245|save 59%|fidelity pass
    

Output Rules

When returning compressed text, output only the compressed file contents. No preamble, no summary, no wrapper fence.

When reporting a completed file workflow, use one line:

signal-compress|target={path}|chars {old}→{new}|est_tok {old}→{new}|save {pct}%|fidelity {pass|fail}

If compression is unsafe:

SIGNAL_DRIFT: <one-line reason>

Fidelity Gate

A compression win counts only when:

  • Every protected token is unchanged.
  • All actionable instructions remain present.
  • Headings stay navigable.
  • Any code, paths, flags, errors, and version numbers remain verbatim.
  • The compressed file can replace the original without changing agent behavior.

Read the full file on GitHub · 88 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 11d ago First seen · 88 lines · 76 tokens per session scan A 88a2c483ad01

Subscribe to this mod's changes

signal-compress is a skill published in the GitHub repository mattbaconz/signal (10 stars, last pushed 2mo ago), licensed MIT. It adds 76 tokens to every session and 698 once invoked, about $0.0004 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.

Related

Other skills, from other repositories

lov-auto-context

Manual or hook-triggered context operator. Evaluates the current session for pollution (long conversations, topic drift, stale noise), AND takes concrete context-shaping actions: writing project memory, updating global or project CLAUDE.md (with diff + confirm), and recommending harness commands like /fork, /compact…

lovstudio/skills · 112 tokens

lov-search-chat

A local search tool for recalling relevant parts of earlier AI conversations from an indexed memory store. It returns matching sessions and identifiable excerpts from the original conversations.

lovstudio/skills · 53 tokens

lov-ataru-indexing

A maintenance tool for building, repairing, or checking a local searchable index of Ataru AI conversation history. An index is an organised lookup structure that makes searching stored records faster.

lovstudio/skills · 55 tokens

lov-memory-add

A knowledge-capture workflow that saves user-selected information as a categorized, tagged note with its source and an index entry. It stores the note as Markdown and can read existing records to avoid duplicates or conflicts.

lovstudio/skills · 43 tokens

lov-memory-search

A search workflow for a configured personal knowledge base, including memory and distill folders. It searches indexes and, when needed, note text, then returns matching titles, excerpts, dates, sources, and real file links.

lovstudio/skills · 42 tokens

learn

Review the current conversation and update project knowledge artifacts - common-gotchas.md (bug patterns), AGENTS.md (conventions), agent memory (cross-session). Use when asked to '/learn', 'what did we learn', 'capture lessons', 'update common-gotchas'.

craigcossairt/trellis · 58 tokens