context-compression

context-compression is a skill for Claude Code from ils15/pantheon-legacy. It costs 32 tokens per session (10,159 once invoked), scanned A, original, MIT.

A system for shortening work summaries while keeping the details most important to each downstream task. It also creates cross-references for items such as endpoints, database tables, migrations, components, and decisions.

In plain words
What is it for?
Condensing task summaries, preserving relevant fields for different agents, managing context budgets, and maintaining references across sprints.
Why use it?
Long-running projects can exceed an agent's available context, causing important details to be lost. Priority-based compression keeps critical information and allocates space according to importance.

Skill for Claude Code

Written for Claude Code: context: fork in frontmatter. Also seen: mentions OpenCode.

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 skills/ils15/pantheon-legacy/context-compression
Any agent
npx skills add ils15/pantheon-legacy --skill context-compression
Clone the repo
git clone --depth 1 https://github.com/ils15/pantheon-legacy

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.

agentmods badge for context-compression

README.md
[![agentmods](https://agentmods.dev/badge/skills/ils15/pantheon-legacy/context-compression.svg)](https://agentmods.dev/skills/ils15/pantheon-legacy/context-compression)
Your own site
<a href="https://agentmods.dev/skills/ils15/pantheon-legacy/context-compression"><img src="https://agentmods.dev/badge/skills/ils15/pantheon-legacy/context-compression.svg" alt="Measured on agentmods" height="20"></a>
Per session 32 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 10,159 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 5.1 $0.00032 $0.10159
Opus 5 $0.00016 $0.05079
Sonnet 5 $0.00006 $0.02032
Haiku 4.5 $0.00003 $0.01016

Measured 6d ago against content hash 07cfddfedeb8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

context-compression 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 6d 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.

.clinerules/skills/context-compression/SKILL.md · 862 lines

How it starts

The opening of the file, as written. The whole thing — 862 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Context Compression — Level 2 (Intelligent Compression)

Level 2 replaces Level 1 with priority-aware compression: scoring, semantic summarization for critical/high entries, downstream-aware field masks, budget allocation, and automatic cross-references. Zeus triggers, Mnemosyne executes.


1. Overview

Level 2 adds four capabilities beyond Level 1:

  1. Priority Scoring Engine — Deterministic keyword-driven scoring (no LLM) for each subtask_summary across 5 dimensions. Outputs CRITICAL/HIGH/MEDIUM/LOW.
  2. Semantic Summarization — Template-generated contextual summaries for CRITICAL/HIGH entries, tailored to the downstream agent pair. MEDIUM/LOW get standard mechanical 1-line compression.
  3. Downstream-Aware Compression — Field masks preserve the most relevant fields per (from_agent, to_agent) pair. Themis gets all fields; Mnemosyne gets minimal.
  4. Cross-Reference Mechanism — Auto-generated _xref/index.md with entity references for endpoints, tables, migrations, components, and decisions, surviving sprint boundaries.

Budget allocation is priority-greedy. CRITICAL entries are always expanded. Overflow escalates to Zeus.


2. Compression Triggers

Same 7 triggers as Level 1, with Level 2 behavior:

Trigger Fires Level 2 Behavior
C1 Phase Gate Themis APPROVED Priority-score subtask_summary → semantic summary (CRITICAL/HIGH) or 1-line (MEDIUM/LOW) → 01-active-context.md
C2 Phase Gate Themis APPROVED IMPL artifact → scored, archived to 02-progress-log.md; .tmp/ deleted
C3 Phase Gate Themis APPROVED REVIEW summary line → 02-progress-log.md; .tmp/ deleted
C4 Feature Complete Last phase APPROVED PLAN reference → 01-active-context.md; .tmp/ deleted
C5 Sprint Close @mnemosyne Close sprint Bulk cleanup + cross-ref archive + flag unresolved
C6 Explicit /compress Per-type compression with priority scoring
C7 Size-based 01-active-context.md Completed Phases > 100 lines Priority-greedy trim: oldest LOW entries first, preserving CRITICAL and HIGH

Read the full file on GitHub · 862 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. 6d ago First seen · 862 lines · 32 tokens per session scan A 07cfddfedeb8

Subscribe to this mod's changes

context-compression is a skill published in the GitHub repository ils15/pantheon-legacy (10 stars, last pushed 3d ago), licensed MIT. It adds 32 tokens to every session and 10,159 once invoked, about $0.0002 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

media-ingest

Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.

garrytan/gbrain · 52 tokens

mem0-oss-to-platform

Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…

mem0ai/mem0 · 273 tokens

Cortex

Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…

danielmiessler/LifeOS · 196 tokens

agent-memory

../../../engineering/agent-memory/skills/agent-memory/SKILL.md.

alirezarezvani/claude-skills · 0 tokens

memory

Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.

opensquilla/opensquilla · 44 tokens

ha-data-stores

Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…

shiwenwen/hope-agent · 115 tokens