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.
npx skills add RockaRhymeLLC/claude-code-skills --skill context-managergit clone --depth 1 https://github.com/RockaRhymeLLC/claude-code-skillsWrote 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/skills/rockarhymellc/claude-code-skills/context-manager)<a href="https://agentmods.dev/skills/rockarhymellc/claude-code-skills/context-manager"><img src="https://agentmods.dev/badge/skills/rockarhymellc/claude-code-skills/context-manager.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00064 | $0.01270 |
| Opus 5 | $0.00032 | $0.00635 |
| Sonnet 5 | $0.00013 | $0.00254 |
| Haiku 4.5 | $0.00006 | $0.00127 |
Grade B, and why
context-manager scanned grade B with 1 finding 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 8d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
cat .claude/state/context-usage.json 2>/dev/null How it starts
The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context Manager
Prevent information loss when your Claude Code context window fills up. Automatically tracks usage, saves state, and restores after restart.
When to Activate
- User says
/ctxor/ctx saveor/ctx check - User asks about context usage or remaining space
- User is working on a long task that might exhaust context
- Before starting a large task (reading many files, big implementation)
The Problem
Claude Code has a finite context window. When it fills up:
- The system compresses earlier messages (lossy)
- You lose working state, partial progress, and nuanced context
- Multi-step tasks break halfway through
- You repeat work you already did
Instructions
/ctx check — Check Current Usage
Read the context-usage.json file (updated by Claude Code hooks):
cat .claude/state/context-usage.json 2>/dev/null
Report to the user:
- Green (>60% remaining): Plenty of room. Work freely.
- Yellow (40-60% remaining): Getting warm. Save state before starting big tasks.
- Red (<40% remaining): Save state now. Consider restarting before continuing.
If no tracking file exists, guide setup (see Setup section).
/ctx save — Save Current State
Capture everything needed to resume after a restart:
- Create state file at
.claude/state/assistant-state.md:
# Assistant State
**Saved**: {ISO timestamp}
**Reason**: {why saving — manual save, low context, task checkpoint}
## Current Task
{What you're working on right now}
## What's Done
{Completed steps, with file paths and commit hashes}
## What's Next
{Remaining steps, in order}
## Key Context
{Important facts that would be lost — variable names, design decisions,
user preferences expressed during this session, error messages that
led to the current approach}
## Open Files
{Files currently being worked on, with relevant line ranges}
- Confirm save — tell the user what was captured
/ctx restore — Restore After Restart
At session start, check for saved state:
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.
- 8d ago First seen · 178 lines · 64 tokens per session scan B 6678bc884924
context-manager is a skill published in the GitHub repository RockaRhymeLLC/claude-code-skills (1 stars, last pushed 6mo ago), licensed MIT. It adds 64 tokens to every session and 1,270 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
self-improve
Extract lessons from the current session, or sweep the project's past sessions when asked, and route them to the appropriate knowledge layer (project AGENTS.md, auto memory, existing skills, or new skills). Use when the user asks to "self-improve", "distill this session", "distill past sessions", "sweep past…
recall-reasoning
Recall the reasoning behind a past change by locating the Claude Code transcript that produced it. Use when the user asks to "recall reasoning", "find reasoning", "look up reasoning", "recall implementation reasoning", "find the rationale", "why did I do X", "recall from transcripts", or "find the transcript for this…
memory-hygiene
A manual tool for sorting saved session memories into Hot, Warm, and Cold groups, and finding memories that may be moved or retired.
session-manager
A fallback skill for recording development sessions and restoring project context from an Obsidian vault, a folder of linked notes. It handles explicit resume, handoff, status, and end-of-session requests when automatic capture is unavailable.
cds-stats
Show project memory statistics: session count, observation counts by type, entity count, top entities, and project planning status.
cds-search
Search past session observations using the sessions.search MCP tool. Falls back to CLI if the MCP server is not available.