A coding agent for adding semantic search and automatic memory capture to Akashic Context, an MCP server that gives AI agents persistent memory.

Use it to implement in-process cosine-similarity search, combine vector and keyword results, and add a memory tool that extracts and removes duplicate memories.

It removes the need to rely only on keyword matching or manually saved memories when building the first search-and-memory features.

sprint-1 — Agent by tostechbr

Install

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.

agentmods

npx agentmods add agents/tostechbr/memoryclaw/sprint-1

Clone the repo

git clone --depth 1 https://github.com/tostechbr/memoryClaw

Made for: Claude Code.

Per session 55 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,153 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00055	$0.01153
Opus 5	$0.00028	$0.00576
Sonnet 5	$0.00011	$0.00231
Haiku 4.5	$0.00006	$0.00115

Measured yesterday against content hash e59226453a5e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

sprint-1 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 yesterday.

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.

.claude/agents/sprint-1.md · 114 lines

How it starts

The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Sprint 1 Agent — Vector Search + memory_add

You are a senior TypeScript engineer implementing Sprint 1 of Akashic Context.

Project Context

Akashic Context is an open-source MCP server that gives AI agents persistent memory. Sprint 0 (multi-user isolation) is COMPLETE. Sprint 1 adds semantic search and auto-extraction.

Stack: TypeScript ESM, Node 18+, better-sqlite3, Vitest, MCP Protocol, OpenAI API Package manager: pnpm (monorepo) Test command: pnpm test -- --run Build command: pnpm build

Sprint 1 Deliverables

1. cosineSimilarity() + searchVectorInProcess()

File: packages/core/src/memory/storage.ts

Add module-level cosineSimilarity(a, b) function
Add searchVectorInProcess(params) method to MemoryStorage
See vector-search-patterns skill for exact implementation

2. Wire hybrid merge with real vector results

File: packages/core/src/memory/manager.ts (and/or hybrid.ts)

searchMemory() currently falls back to keyword-only when vecAvailable = false
Change fallback to use searchVectorInProcess() instead
Now hybrid merge gets REAL vector results

3. memory_add MCP tool

File: packages/mcp-server/src/index.ts

New tool: memory_add({ message, userId? })
Pipeline: embed → searchVectorInProcess → LLM merge OR extract → memory_store
Add callLLM(prompt) private method using OpenAI chat completions
Add extractionModel?: string to server config (default: "gpt-4o-mini")
See memory-add-patterns skill for exact implementation

4. Tests (minimum 8 new tests)

Files: packages/core/src/memory/storage.test.ts + packages/mcp-server/src/index.test.ts

cosineSimilarity mathematical tests (identical=1, orthogonal=0, opposite=-1)
searchVectorInProcess returns results sorted by similarity
memory_add creates new file when no similar found
memory_add merges when similar found (similarity >= 0.85)
memory_add different topics = separate files (no false merge)

Design Decisions (already made — do not re-debate)

Read the full file on GitHub · 114 lines

Changes

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.

yesterday First seen · 114 lines · 55 tokens per session scan A e59226453a5e

Subscribe to this mod's changes

sprint-1 is an agent published in the GitHub repository tostechbr/memoryClaw (8 stars, last pushed 5mo ago), licensed MIT. It adds 55 tokens to every session and 1,153 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.

Git and workflow LLM Memory and context Project management agent context-engine mcp mcp-server

project-manager-backlog

Use this agent when you need to manage project tasks using the backlog.md CLI tool. This includes creating new tasks, editing tasks, ensuring tasks follow the proper format and guidelines, breaking down large tasks into atomic units, and maintaining the project's task management workflow. Examples: Context: User wants…

MrLesk/Backlog.md · 0 tokens

f1-test-drive

Orchestrate F1 test drives to validate the Cyrus agent system end-to-end. Use this agent to run comprehensive test drives that verify issue-tracker, EdgeWorker, and renderer components.

cyrusagents/cyrus · 43 tokens

lead

Use this agent to orchestrate a full task from the harness backlog: decompose it into a plan, delegate to explorer, builder, and reviewer in sequence, and close the session correctly. Invoke when starting a new work session, picking up a pending task, or when another agent reports a blocker that requires…

enmanuelmag/agent-harness-kit · 67 tokens

sprint-1