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 troyjthomas/ios-agent-skills --skill session-managementgit clone --depth 1 https://github.com/troyjthomas/ios-agent-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/troyjthomas/ios-agent-skills/session-management)<a href="https://agentmods.dev/skills/troyjthomas/ios-agent-skills/session-management"><img src="https://agentmods.dev/badge/skills/troyjthomas/ios-agent-skills/session-management/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/skills/troyjthomas/ios-agent-skills/session-management"><img src="https://agentmods.dev/badge/skills/troyjthomas/ios-agent-skills/session-management.svg" alt="Reviewed on agentmods" width="80" 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 | $0.01787 |
| Opus 5 | $0.00032 | $0.00894 |
| Sonnet 5 | $0.00013 | $0.00357 |
| Haiku 4.5 | $0.00006 | $0.00179 |
Grade A, and why
session-management 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 208 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Session Management
Claude Code has no memory between sessions. Every new session starts blank and reads your CLAUDE.md. This skill ensures continuity.
Context Window Management
Claude Code's context window is large but finite. Everything loaded into it competes for attention. Overload it and quality drops.
The 500-Line Rule
Your CLAUDE.md should stay under 500 lines. This is the file loaded into every session. If it's too long, Claude Code starts ignoring rules at the bottom.
What belongs in CLAUDE.md (always loaded):
- Core principles (native SwiftUI, iOS version, HIG)
- Navigation structure
- Architecture conventions
- Custom components list
- What NOT to do
- Known gotchas
- Build configuration
- .pbxproj safety rule
What belongs in separate files (loaded when needed):
- App spec (APP_SPEC.md)
- Design system (DESIGN_SYSTEM.md)
- Detailed component specs
- Reference documentation
Per-Session Context Loading
At the start of each session, tell Claude Code what to focus on:
Read CLAUDE.md. Today we're working on the Settings flow.
Also read the Settings section of APP_SPEC.md. Ignore other
sections for now.
This focuses the context on what matters for THIS session rather than loading everything.
Signs of Context Overload
- Claude Code starts ignoring rules from your CLAUDE.md
- It builds custom components when native ones exist
- It forgets the architecture pattern mid-session
- Responses get slower or less precise
- It repeats mistakes you've already corrected
Fix: Start a new session. Load only the CLAUDE.md and the specific section you're working on. Keep the scope narrow.
Session Continuity
The Problem
Claude Code sessions don't survive Mac restarts, long periods of inactivity, or network timeouts. When a session ends, all conversation context is lost. The code on disk is fine (it's committed to git), but Claude Code doesn't know what it was doing.
The Solution: Git as Your Checkpoint System
Your git history IS your session memory. Every commit message tells the next session what happened.
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.
- 9d ago First seen · 208 lines · 64 tokens per session scan A 57f725096e89
session-management is a skill published in the GitHub repository troyjthomas/ios-agent-skills (2 stars, last pushed 3mo ago), licensed MIT. It adds 64 tokens to every session and 1,787 once invoked, about $0.0003 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.
Other skills, from other repositories
context-engineering
A project documentation framework that records a project's goals, requirements, roadmap, and current state in separate files. It also keeps those documents within set size limits for an AI coding agent.
error-experience-library
An error knowledge base that stores recurring error patterns, their causes, and the fixes that worked. It can search past entries and track whether each suggested fix succeeded.
mnemos
Task-scoped memory lifecycle — typed MnemoGraph prevents lossy context compaction by treating facts/decisions/code-refs/handoffs as distinct node types with per-type eviction policies.
durable-session-state
Persist plans, scope decisions, evidence, and reviewer/critic verdicts to durable files during long or multi-phase tasks so work survives context compaction, session resumes, and handoffs. Use for swarm-mode tasks, before context grows large, when recording approval gates, and when resuming after compaction or a…
token-optimization
Use when the user says 'token optimization', 'save tokens', 'context window', 'reduce tokens', 'token stack', or 'TokenStack', or asks about extending context window capacity. Covers TokenStack, the built-in compression proxy that shrinks Claude Code tool output before it reaches the Anthropic API. Do NOT use for…
grimoire
Use when the user says 'update context', 'update claude', 'save library', or after significant project changes.