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 anhnguyen0905/codex-mcp --skill agent-context-persistencegit clone --depth 1 https://github.com/anhnguyen0905/codex-mcpWrote 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/anhnguyen0905/codex-mcp/agent-context-persistence)<a href="https://agentmods.dev/skills/anhnguyen0905/codex-mcp/agent-context-persistence"><img src="https://agentmods.dev/badge/skills/anhnguyen0905/codex-mcp/agent-context-persistence.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 | $0.00069 | $0.00725 |
| Opus 5 | $0.00034 | $0.00362 |
| Sonnet 5 | $0.00014 | $0.00145 |
| Haiku 4.5 | $0.00007 | $0.00072 |
Grade A, and why
agent-context-persistence 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 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.
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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Context Persistence (embed into Codex prompts)
How to persist agent working state (plans, decision logs, task files) so a fresh session on a LARGE
project resumes within a fixed token budget without trusting stale claims. Distilled from
docs/context-persistence-research-2026-08-06.md (Anthropic context-engineering, Manus, Cognition,
Letta/MemGPT, Aider repo-map, spec-kit, event-sourcing); items marked (derived) are our own
inference, verify in review.
Standards block
Context-persistence rules (mandatory when writing or reading agent state files):
- Fixed read budget: every read-back artifact has a hard token ceiling enforced by code
(rank items, cut to budget), never by prose convention. Budget overflow drops whole
lowest-priority items, not truncated sentences.
- Restorable compression: when something is dropped from a slice, append a pointer line naming
where the full content lives (file + section). Content is one read away; never silently omit.
- Pointers, not payloads: durable state references code as path:line + enclosing symbol,
never pasted snippets — pasted code is stale the moment the file changes.
- Anchor every write: state blocks record the git HEAD SHA at append time. On read-back,
git diff --stat <anchor>..HEAD plus a dirty-tree check classifies each block fresh vs verify;
a block without an anchor is always verify.
- Trust but verify: written state is hypothesis; filesystem + git are ground truth. A resuming
agent must re-check any [verify] claim against current code before acting on it.
- Deterministic over generated: derived views (slices, resume summaries) are produced by
deterministic code from the source-of-truth file, regenerated on every read — never by an LLM
summarization step (which silently loses constraints) and never hand-edited.
- Single source of truth: the append-only log/plan file remains authoritative on disk; derived
slice files are disposable and regenerable at any time.
- Never persist a code inventory: persist "where to look" (contracts, entry points, pointers);
regenerate any structural view of the code at read time. Persisted descriptions of code are
drift source #1. (derived from Claude Code no-index finding + Aider rebuild-per-message)
- Budget is a test invariant: tests build a large fixture (long log, many tasks) and assert
tokensOf(slice) <= budget, plus a legacy-format fixture that degrades without crashing.
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 · 50 lines · 69 tokens per session scan A e1c1e175facc
agent-context-persistence is a skill published in the GitHub repository anhnguyen0905/codex-mcp (3 stars, last pushed 5d ago), licensed MIT. It adds 69 tokens to every session and 725 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
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.
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.…
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…
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.
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…
establishing-project-context
Use when the user asks to establish shared project language, or project work exposes a conflicting, renamed, or deprecated domain term that needs active semantic modeling. Routine small tasks stay on the fast path.