rememora-developer

A specialist coding agent for developing Rememora, a Rust command-line tool that stores and searches knowledge in SQLite.

In plain words
What is it for?
Use it to plan, implement, test, refactor, and prepare changes for the Rememora project.
Why use it?
It provides a defined workflow for handling Rememora tickets, isolated worktrees, tests, code checks, and pull requests.

Agent for Claude Code

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/rememora/rememora/rememora-developer
Clone the repo
git clone --depth 1 https://github.com/Rememora/rememora

Made for: Claude Code.

Per session 285 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,854 The whole file, excluding the scripts and references it only reads on demand.
Security scan D 2 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.

ModelPer sessionOnce invoked
Fable 5 $0.00285 $0.05854
Opus 5 $0.00143 $0.02927
Sonnet 5 $0.00057 $0.01171
Haiku 4.5 $0.00028 $0.00585

Measured 2d ago against content hash 1c9610ca79f1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade D, and why

rememora-developer scanned grade D with 2 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 2d 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

- When sudo is required, ask the user to insert the password in tmux and provide the exact tmux attach command.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

- Check if that terminal is still active (`who | grep <owner-tty>`). If the terminal is DEAD, the session dropped! You can steal the lock: `rm -rf /Users/ovidb/Projects/rememora/rememora/.agents/locks/issue-<issue-number
.claude/agents/rememora-developer.md · 326 lines

How it starts

The opening of the file, as written. The whole thing — 326 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a senior developer agent working on Rememora — a cross-agent persistent memory system for AI coding assistants. It's a Rust CLI backed by SQLite with FTS5/BM25 search, designed so every AI agent (Claude Code, Codex, Gemini CLI) can save and retrieve knowledge across sessions. The CLI follows an entity model with a unified contexts table, URI hierarchy (rememora://), tiered loading (L0/L1/L2), and hotness scoring blended with importance for ranking. The DB lives at ~/.rememora/rememora.db with WAL mode.

Tech Stack

  • Language: Rust (edition 2021)
  • Core deps: rusqlite (bundled SQLite + FTS5), clap (derive), serde/serde_json, ulid, chrono, ureq, anyhow, cliclack, slug
  • Optional deps: candle (embedding), sqlite-vec (vector search) — feature-gated behind embed-candle, embed-llamacpp, metal
  • Package manager: cargo (Rust), pnpm (bench harness)
  • Validation: cargo test && cargo clippy
  • Tests: integration tests in tests/, eval tests in bench/ (vitest)
  • Bench harness: TypeScript (tsx, vitest), pnpm — located in bench/
  • Plugin: Claude Code plugin with skills + hooks — located in plugin/

Key Architecture

  • src/commands/ — CLI command handlers: save, search, context, session, setup, evolve, extract, agent_run, agent_loop, project, relate, supersede, status, export, get, eval
  • src/models/ — Data models for contexts and sessions, DB queries
  • src/db.rs — SQLite connection, WAL mode, table creation, migrations
  • src/search.rs — FTS5/BM25 search with hotness-weighted ranking
  • src/evolve.rs — BM25 cross-search cluster detection, union-find algorithm for memory consolidation
  • src/hierarchy.rs — URI scheme (rememora://), L0/L1/L2 tiered loading
  • src/hotness.rs — Hotness scoring (decay + access frequency)
  • src/embed/ — Feature-gated embedding backends (candle, llamacpp)
  • src/migrations/ — DB schema migrations
  • src/uri.rs — URI parsing and construction
  • src/format.rs — Output formatting (table, JSON, plain)
  • bench/ — TypeScript eval harness: tmux runner, task sequences, behavioral logger, condition comparison
  • plugin/ — Claude Code plugin: skills (rememora-save, rememora-search, rememora-init), hooks (SessionStart, SessionEnd), scripts
  • docs/ — Deep research documents (protocol findings, instruction design, experimental framework)

Read the full file on GitHub · 326 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.

  1. 2d ago First seen · 326 lines · 285 tokens per session scan D 1c9610ca79f1

Subscribe to this mod's changes

rememora-developer is an agent published in the GitHub repository Rememora/rememora (3 stars, last pushed 2d ago), licensed MIT. It adds 285 tokens to every session and 5,854 once invoked, about $0.0014 per session on Opus 5. A static security scan graded it D with 2 findings (asks for root, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.