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 agentmods add rules/alieismy/codex-three-layer-delivery/04-context-sessiongit clone --depth 1 https://github.com/alieismy/codex-three-layer-deliveryWrote 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/rules/alieismy/codex-three-layer-delivery/04-context-session)<a href="https://agentmods.dev/rules/alieismy/codex-three-layer-delivery/04-context-session"><img src="https://agentmods.dev/badge/rules/alieismy/codex-three-layer-delivery/04-context-session.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 | $0.00616 | $0.00616 |
| Opus 5 | $0.00308 | $0.00308 |
| Sonnet 5 | $0.00123 | $0.00123 |
| Haiku 4.5 | $0.00062 | $0.00062 |
Grade A, and why
04-context-session 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 4d 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context And Session Management
Context Health
- Use full file paths in plans, review reports, and final reports.
- Summarize key constraints before complex multi-step work.
- Refresh context when the task changes between requirements, feasibility, research, proposal, detailed design, standards work, professional writing, review, and delivery orchestration.
- If answer quality degrades or the same issue fails twice, stop and reassess context before continuing.
- Reuse previously inspected stable project context only when its authoritative source, revision, and applicability remain known.
- Before substantive work, use applicable low-cost change detectors such as branch and HEAD, worktree status, changed paths, and relevant source revisions to scope the refresh; do not rescan the entire repository by default.
- Re-read affected or decision-critical sources after revision changes, external edits, new evidence, context compaction, task redirection, or conflicting observations. Do not claim prior context remains current when applicable change detectors have not been checked.
- For complex or high-impact work, briefly identify the reused baseline and newly refreshed dynamic evidence at the first meaningful checkpoint when that improves reviewability; do not impose a fixed context-accounting format on simple tasks.
Task State Boundary
- Keep current goals, progress, branch or commit snapshots, test results, known issues, failed approaches, and next actions in a task plan or delivery record, not in this file. This file may define when and where to read that state but must remain limited to durable project rules.
Session Strategy
For large tasks, split work into independently verifiable sessions:
| Session | Goal | Persistence |
|---|---|---|
| 1 | Understand documents, scope, and constraints | Notes, rules, or document map |
| 2 | Draft or revise document content | Tracked document changes |
| 3 | Independent review and verification | Review findings and verification notes |
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.
- 4d ago First seen · 60 lines · 616 tokens per session scan A 13fb03afbded
04-context-session is a cursor rule published in the GitHub repository alieismy/codex-three-layer-delivery (2 stars, last pushed 4d ago), licensed MIT. It adds 616 tokens to every session, about $0.0031 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 cursor rules, from other repositories
mempalace-recall-always
Always-on MemPalace recall — search the palace before answering about past work, people, projects, or prior decisions.
dreamd-recall
Recall lessons, decisions, and prior context from the .agent/ memory daemon. Use when starting work in a project that has a .agent/ folder, when the user references a past decision, or when you are about to make a choice that has a documented prior.
session-memory
Use at conversation wrap-up or when the user explicitly indicates end-of-session — capture residual lessons not captured in-flight.
common_memory_bank
I am Cursor, an expert software engineer with a unique characteristic: my memory resets completely between sessions. This isn't a limitation - it's what drives me to maintain perfect documentation. After each reset, I rely ENTIRELY on my Memory Bank to understand the project and continue work effectively. I MUST read…
context-recorder-system
Context Recorder System (记录员系统) - 模块化索引文件.
self-improving-obsidian-llm-wiki
LLM Wiki OS operating rules.