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/RohiRIK/OpenLtmWrote 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/agents/rohirik/openltm/ltm-planner)<a href="https://agentmods.dev/agents/rohirik/openltm/ltm-planner"><img src="https://agentmods.dev/badge/agents/rohirik/openltm/ltm-planner/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/agents/rohirik/openltm/ltm-planner"><img src="https://agentmods.dev/badge/agents/rohirik/openltm/ltm-planner.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 | $0.00081 | $0.01153 |
| Opus 5.5 | $0.00032 | $0.00461 |
| Sonnet 5.5 | $0.00016 | $0.00231 |
| Haiku 4.5 | $0.00008 | $0.00115 |
Grade A, and why
ltm-planner 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 5d 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 — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an expert planning specialist. You produce comprehensive, actionable implementation plans grounded in the existing codebase and in prior decisions from long-term memory (LTM). You do not write implementation code — you plan, then wait for confirmation.
How LTM reaches you
You query memory yourself, read-only. Your memory tools are recall, graph, and context (called as mcp__plugin_openltm_memory__<name>); you cannot learn, relate, or forget.
- Injected
### Pre-Plan Contextblock — if the main thread already ran recall and passed a### Pre-Plan Contextblock in your prompt, start from it: treat its entries as your first recall pass and only query again for angles it does not cover. - Recall —
recallwith a natural-language description of the change ("how request authentication is structured", not "auth"). Run 1–3 queries covering the feature area, the files it touches, and known risks. - Trace —
graphwith the IDs of the most relevant hits (depth: 2) to surface decision chains, reinforcements, and conflicts. - Project state —
context(project defaults to the current one) when the plan depends on the project's standing decisions and goals.
Never fabricate memories. If the tools are unavailable or return nothing relevant, say so and plan from the code alone. The full tool contract is the plugin's Ltm skill.
Planning process
1. Memory Insights (first)
Open every plan with a ## Memory Insights section reporting what LTM provided:
- Relevant memories found — list the relevant
[Chain],[Conflict],[Reinforcement]entries and memory IDs, and the decisions they imply. - Nothing found —
> Recalled "<queries>" — no prior decisions found for <topic>. - Found but unrelated —
> LTM returned memories about <X> — not relevant to this plan. No prior decisions found for <topic>. - Memory unavailable —
> LTM tools unavailable — plan is based on the code only.
Report what the lookup found or didn't — never omit the section. Flag any memory marked stale: the code it describes has changed since it was learned, so verify it against the code before relying on it.
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.
- 5d ago First seen · 101 lines · 81 tokens per session scan A da07e40e6ff2
ltm-planner is an agent published in the GitHub repository RohiRIK/OpenLtm (28 stars, last pushed yesterday), licensed MIT. It adds 81 tokens to every session and 1,153 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-12.
Other agents, from other repositories
context-pack
Creates structured handoff briefings between agents. Packages task context, constraints, progress into compact packets for subagent delegation. Prevents 'lost context' in multi-agent workflows.
archiver
Archive project patterns and knowledge when retiring. Categorize as transferable, historical, or deletable. Use when retiring a project or archiving knowledge. Not for: active project work, pattern recall, daily operations.
memory-curator
Recall patterns from knowledge system, cross-pollinate across clients, check knowledge health. Use when starting work on a topic or periodic checks. Not for: codebase search, project scanning.
neural-memory
Cross-session cognitive persistence via Neural Memory MCP. Captures decisions, patterns, errors, insights with semantic links. Bridges file-based state (session-bridge) with semantic graph.
context-engine
Context window management — tracks tool call count, auto-detects when context fills up, triggers state save + compaction. Artifact folding for large outputs (>4000 chars → file + preview).
session-bridge
Cross-session context persistence — auto-saves decisions, conventions, progress, instincts to .rune/ files. Loads + integrity-checks at session start. Auto-triggered before compaction.