Beads
A memory management tool for coding agents from the Gastown ecosystem that replaces scattered draft notes with a structured issue ledger. Each state change is recorded as an issue carrying an ID, status, and parent-child relationships, with a CLI interface and local data storage.
Timeline (last 90 days)
PentaCore is not suited to dependency-based software development task tracking (Beads or Claude Code's built-in tasks should be used), short tasks within a single session, or long-term memory about the user or the user's preferences.
The Beads architecture redefines context as a structured, navigable graph of 'beads' rather than a flat token stream.
The Beads engine acts as a context router: when the agent needs to perform a task, the system queries the Bead graph for relevant Beads, sorts them by relevance and recency, and injects only the top-N Beads into the prompt.
Beads supports hierarchical context management: high-level Beads include architecture diagrams and module dependencies, while low-level Beads include concrete function implementations and variable declarations, and the agent zooms in or out by traversing the Bead graph.
Because Beads are stored externally, the agent's memory can persist across sessions, enabling it to continue previous work days later while retaining awareness of past decisions and constraints.
Implementing a Bead system adds complexity, requiring management of indexing pipelines, graph relationships and retrieval logic, and it is not a plug-and-play replacement for a simple prompt template.
Incremental indexing and lazy loading are the core strategies for mitigating the cost of building graphs for large-scale codebases, making Beads feasible in large monorepos, but lazy loading brings trade-offs at subgraph boundaries.
The Beads architecture is especially powerful in multi-agent scenarios, where different agents can share the same Bead graph, enabling collaborative memory and specialized Beads, turning coding agents into nodes in a collective intelligence network.
Compared with traditional linear context models, the Beads graph model has advantages in context window size, memory persistence, relevance ranking, navigation method and cost.
Beads, from the Gastown ecosystem, replaces an agent's draft notes with a structured issue ledger, writing each state change as an issue with an ID, status, and parent-child relationships.
All Facts (10)
PentaCore is not suited to dependency-based software development task tracking (Beads or Claude Code's built-in tasks should be used), short tasks within a single session, or long-term memory about the user or the user's preferences.
50%UnverifiedThe Beads architecture redefines context as a structured, navigable graph of 'beads' rather than a flat token stream.
50%UnverifiedThe Beads engine acts as a context router: when the agent needs to perform a task, the system queries the Bead graph for relevant Beads, sorts them by relevance and recency, and injects only the top-N Beads into the prompt.
50%UnverifiedBeads supports hierarchical context management: high-level Beads include architecture diagrams and module dependencies, while low-level Beads include concrete function implementations and variable declarations, and the agent zooms in or out by traversing the Bead graph.
50%UnverifiedBecause Beads are stored externally, the agent's memory can persist across sessions, enabling it to continue previous work days later while retaining awareness of past decisions and constraints.
50%UnverifiedImplementing a Bead system adds complexity, requiring management of indexing pipelines, graph relationships and retrieval logic, and it is not a plug-and-play replacement for a simple prompt template.
50%UnverifiedIncremental indexing and lazy loading are the core strategies for mitigating the cost of building graphs for large-scale codebases, making Beads feasible in large monorepos, but lazy loading brings trade-offs at subgraph boundaries.
50%UnverifiedCompared with traditional linear context models, the Beads graph model has advantages in context window size, memory persistence, relevance ranking, navigation method and cost.
50%UnverifiedBeads, from the Gastown ecosystem, replaces an agent's draft notes with a structured issue ledger, writing each state change as an issue with an ID, status, and parent-child relationships.
50%UnverifiedThe Beads architecture is especially powerful in multi-agent scenarios, where different agents can share the same Bead graph, enabling collaborative memory and specialized Beads, turning coding agents into nodes in a collective intelligence network.
50%