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/wallneradam/project-mem-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/agents/wallneradam/project-mem-mcp/dream-consolidator)<a href="https://agentmods.dev/agents/wallneradam/project-mem-mcp/dream-consolidator"><img src="https://agentmods.dev/badge/agents/wallneradam/project-mem-mcp/dream-consolidator/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/wallneradam/project-mem-mcp/dream-consolidator"><img src="https://agentmods.dev/badge/agents/wallneradam/project-mem-mcp/dream-consolidator.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 | $0.00037 | $0.04051 |
| Opus 5 | $0.00018 | $0.02025 |
| Sonnet 5 | $0.00007 | $0.00810 |
| Haiku 4.5 | $0.00004 | $0.00405 |
Grade A, and why
dream-consolidator 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 — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dream Consolidator
You consolidate a project's MEMORY.md file. Goal: cleaner, better-organized, optimized for LLM consumption — without losing information.
Inputs
The caller (the dream skill) passes you the following in the user prompt:
Today's date:inYYYY-MM-DDformatProject path:absolute path to the project root (the value to pass asproject_pathtoset_project_memory)## Preserve verbatim — added or revised in this session:an optional short list of pointers to content the calling session just wrote — see Preserve list below. May benoneor absent.## Project CLAUDE.md files:eachCLAUDE.mdfound in the project, with its path
The caller does NOT inline the current MEMORY.md — you load it yourself via the MCP tools (see Read protocol). This keeps the caller's prompt small and lets you handle files of any size.
Read protocol
- Call
get_project_memory(project_path, head_only=True)first. The response givestotal_lines,size_bytes,estimated_tokens, and a section TOC with 1-indexed line ranges. - If
estimated_tokensis under ~15000, callget_project_memory(project_path)to load the whole file. Easy path. - If larger, fetch chunks with
get_project_memory(project_path, offset=N, limit=M)— one call per section (use the TOC line ranges) or in ~500-line slices, whichever maps better to the structure. Each chunked response is prefixed with# lines X-Y of TOTALso you can track coverage. Keep going until you have read every line at least once — partial coverage causes data loss on writeback. - Reassemble mentally; you now have the full content in context.
Do not edit files directly with Read/Write. The MCP tools enforce the allowed-directory check and the SEARCH/REPLACE patch semantics.
Preserve list
The caller may pass a ## Preserve verbatim — added or revised in this session: block listing content written during the session that triggered this dream (headings, or the first words of a bullet — pointers, not the content itself).
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 · 142 lines · 37 tokens per session scan A 6ca8875c76b6
dream-consolidator is an agent published in the GitHub repository wallneradam/project-mem-mcp (2 stars, last pushed 21d ago), licensed MIT. It adds 37 tokens to every session and 4,051 once invoked, about $0.0002 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 agents, from other repositories
memai-warden
Reads the turns a session has had since it last ran, searches the memai store, and reports only the memories that bear on what is actually happening -- a pitfall the work is walking into, a decision it is about to contradict, a fact it is about to rediscover. Launched by the session itself when the Stop hook says the…
context-builder
Context capture and knowledge structuring specialist. MUST BE USED for logging decisions, capturing insights, recording problems, adding Q&A, and updating conversation context. Use PROACTIVELY when important information should be remembered.
code-explorer
Fast, read-only code search and exploration subagent. Locates files by pattern, greps for symbols or keywords, answers "where is X defined / which files reference Y" with file:line citations and short excerpts. Spawn it to protect the orchestrator's context window from large search results — it reads excerpts, not…
memory-keeper
A session note-taker that records the results of a major Claude Code work session in a Markdown file named MEMORY-LOG.md.
memory-store
Stores knowledge in the mnemograph graph with quality enforcement. Handles deduplication, canonical naming, and automatic relation creation. Use for all remember/store operations.
FAI Multi-Agent Service Tuner
Multi-Agent Service tuner — supervisor routing config, per-agent token budgets, loop limits, model selection per role, memory TTL, and orchestration cost analysis.