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 thammarongg/blueocean-vector --skill blueocean-memorygit clone --depth 1 https://github.com/thammarongg/blueocean-vectorWrote 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/thammarongg/blueocean-vector/blueocean-memory)<a href="https://agentmods.dev/skills/thammarongg/blueocean-vector/blueocean-memory"><img src="https://agentmods.dev/badge/skills/thammarongg/blueocean-vector/blueocean-memory/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/thammarongg/blueocean-vector/blueocean-memory"><img src="https://agentmods.dev/badge/skills/thammarongg/blueocean-vector/blueocean-memory.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 | $0.00106 | $0.00916 |
| Opus 5 | $0.00053 | $0.00458 |
| Sonnet 5 | $0.00021 | $0.00183 |
| Haiku 4.5 | $0.00011 | $0.00092 |
Grade A, and why
blueocean-memory 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 3d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BlueOcean Memory
A shared, persistent memory reachable via the blueocean MCP server
(http://localhost:8765/mcp) -- the same server every registered coding
tool on this machine connects to. Memory written by one tool is readable by
any other; it survives switching agents entirely or running out of tokens
in one tool.
At the start of a session
- Determine
project-- use the current working directory's folder name unless the user says otherwise. - Call
memory_manifest(project)to see what areas/modules already have recorded knowledge. If the project has no memory yet, skip to working normally. - Call
memory_search(project, query="current state / recent decisions", area=<relevant area>)scoped to whichever area(s) look relevant to the task at hand. Read thesummarylayer first; only fetchmemory_getfor a specific entry's full content when the summary isn't enough.
Do this BEFORE assuming a project is unfamiliar or asking the user to re-explain context that may already be recorded.
Every time you decide something
Call memory_store(project, area, module, content, summary, importance)
at the moment a decision is reached, while the reasoning is still in
front of you -- not later, and never saved up for the end of the session.
The trigger is a decision a future session would otherwise have to work out again: why an approach was chosen over the alternatives, what an investigation concluded, what turned out not to work.
- importance=5: architecture decisions, why something was chosen over alternatives, anything a future agent would otherwise have to re-derive.
- importance=3: routine facts, current status, in-progress notes.
- importance=1-2: transient/low-value notes (safe to prune later).
There is no quota. A session that decided nothing writes nothing, and that is correct rather than a miss.
Keep area/module consistent with what memory_manifest already showed
for this project rather than inventing new names for the same concept.
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.
- 3d ago First seen · 84 lines · 106 tokens per session scan A 63c0c15a41f3
blueocean-memory is a skill published in the GitHub repository thammarongg/blueocean-vector (0 stars, last pushed 3d ago), licensed MIT. It adds 106 tokens to every session and 916 once invoked, about $0.0005 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 2026-09-07.
Other skills, from other repositories
scout
Research the web for solutions, competitors, alternatives, or prior art for a decision facing this project — then store the findings in project memory so they survive the session. Use when choosing a library or approach, comparing tools, checking what others do, or evaluating whether to build vs adopt.
recall
Answer "why" and "when" questions about a project's decisions — the intent behind a past choice, when it was made, what was tried before and rejected. Use when asked why something was decided, what the rationale was, or before reversing or redoing earlier work. For why the code mechanically behaves a certain way use…
hermes-history-ingest
Ingest Hermes agent history into the Obsidian wiki. Use this skill when the user wants to mine their past Hermes sessions for knowledge, import their /.hermes folder, extract insights from previous Hermes conversations, or says things like "process my Hermes history", "add my Hermes memories to the wiki", "ingest…
wiki-narrate
Turn a wiki topic into a cited Markdown briefing, plain-language explanation, or progressive lecture. Use this skill for topic-based briefing, explanation, and lecture requests that must stay within the evidence compiled in an Obsidian vault.
session-brain
Build and maintain a topic graph over your agent session history. Reads every Claude session transcript plus the pruned sessions that survive only in history.jsonl, clusters them by topic using local TF-IDF (no API calls, no embeddings), and writes an interactive graph you can open in a browser. Use when the user says…
wiki-context-pack
Produce a token-bounded, citation-ready context slice from an existing Obsidian vault for a downstream agent or task. Use for "/wiki-context-pack", "use my vault as context", "context slice for X", "pack the wiki for my agent", or "bounded context for Y".