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.
git clone --depth 1 https://github.com/cdeust/session-optimizerWrote 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.
[](https://agentmods.dev/plugins/cdeust/session-optimizer/context-guard)<a href="https://agentmods.dev/plugins/cdeust/session-optimizer/context-guard"><img src="https://agentmods.dev/badge/plugins/cdeust/session-optimizer/context-guard/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.
<a href="https://agentmods.dev/plugins/cdeust/session-optimizer/context-guard"><img src="https://agentmods.dev/badge/plugins/cdeust/session-optimizer/context-guard.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
context-guard 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 19d 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.
What it actually says
{
"name": "context-guard",
"description": "Context-budget guard for long Claude Code and Codex sessions: a Stop hook enforces a per-model checkpoint protocol (mechanical stub + budgeted memory-writer subagent on Claude; a direct stub-write instruction on Codex), and a SubagentStop tracker surfaces true session spend (main thread + subagents) on Claude. Default protocol uses only vanilla Claude Code; a scoped memory layer is detected at runtime when installed.",
"version": "2.1.0",
"author": {
"name": "Clement Deust",
"email": "[email protected]"
},
"homepage": "https://github.com/cdeust/session-optimizer",
"repository": "https://github.com/cdeust/session-optimizer",
"license": "MIT",
"keywords": [
"claude-code",
"codex",
"context-window",
"token-budget",
"checkpoint",
"stop-hook",
"subagent-cost",
"session-management"
]
}
What it installs
The manifest is a name and a version. 1 agent travel with it, and installing the plugin installs all of them — 298 tokens a session between them. Each is measured on its own page, and each can be installed alone.
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.
- 19d ago Changed · +1 lines 11c3373d9c16
- 1mo ago First seen · 22 lines scan A 35f5c1a8012b
context-guard is a plugin published in the GitHub repository cdeust/session-optimizer (1 stars, last pushed 2d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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.
Other plugins, from other repositories
claude-brain
Sync and evolve your Claude Code brain across machines. Automatically exports, merges, and applies your accumulated knowledge (memory, skills, agents, rules, settings) across all your development machines via Git.
cc-reload
Session continuity across context resets for Claude Code. Snapshots a session digest before /clear, /compact, or auto-compaction and auto-rehydrates it after. The non-looped companion to cc-repete — stands down whenever a cc-repete loop is active.
recall
Durable, fully-local project memory for Claude Code. Logs every session to .recall/history.md and condenses it into .recall/context.md with a local Python summarizer (TF-IDF + TextRank) — no API key, no external model, nothing leaves your machine. New sessions resume from the saved context.
claude-code-hermit
Your own local Claude Tag, in the Claude app, Discord, Telegram, or a custom Claude Code channel.
howl
Your AI screams — Howl listens. Blazing-fast Go statusline HUD with adaptive layouts and per-metric control.
super-bootstrap
Bootstrap or sync the development pipeline in any repo. /super-bootstrap:setup is a thin orchestrator: it installs/syncs the generic harness runway (CLAUDE.md, AGENTS.md foreign-executor contract, skeleton docs, path-scoped rules, the core plugin pin, default-on harness hooks (commit-channel routing raw commits to the.