Continuity - the work graph for coding agents
No more side quests, forgotten context, or vibes where receipts should be. Continuity gives coding agents one outcome, living evidence, real blockers, and the next safe task.
Get started
Set it up yourself, or copy an agent-ready brief into Codex, Claude Code, Cursor, or your runner. Either path uses the public CLI, one workspace token, and the hosted graph.
Install the source-free package, export your token, then verify the hosted workspace.
npm install -g @ligence/continuity
export CONTINUITY_API_TOKEN=<workspace-token>
continuity setup --json
continuity doctor --jsonPaste a short brief into your coding agent so it installs the CLI, verifies setup, asks for the next graph packet, and reconciles the run.
Copies Markdown for Codex, Claude Code, Cursor, or another runner.
# Install Continuity
Install the hosted Continuity CLI and connect this workspace.
1. Install the CLI:
npm install -g @ligence/continuity
2. Ask the user for CONTINUITY_API_TOKEN if it is not already set.
3. Export the token:
export CONTINUITY_API_TOKEN=<workspace-token>
4. Run setup and verification:
continuity setup --json
continuity doctor --json
5. Pull your first work packet:
continuity next --agent codex --horizon overnight --compact --json
6. Before doing work, inspect the assigned node:
continuity agent-context --node <node-id> --agent codex --json
7. After work, reconcile the run:
continuity report-delta --node <node-id> --kind completed --summary "<what changed>" --evidence "<proof>" --json
Rules:
- Do not ask for CONTINUITY_WORKSPACE_ID.
- Do not set CONTINUITY_API_URL unless using staging.
- If you are not Codex, replace codex in --agent flags with your agent name.
- Treat the graph as the source of truth for goals, blockers, evidence, and next action.Orientation beats drift
An unguided agent drifts across competing signals until a human redirects it.
The agent starts from a prompt, follows several fading traces toward side quests and stale context, then needs a human redirect before reaching partial output.
prompt agent side quest stale thread unfiled debt partial output human redirectThe agent follows whatever signal is loudest. It can self-steer into side quests until a human re-enters the loop.
A work graph connects the active goal to tasks, decisions, evidence, debt, and the next action.
The graph starts with the active goal, branches into a decision and implementation task, reconciles evidence and debt, then selects the next runnable task.
Continuity gives every run provenance and direction: where it came from, what changed, and the next safe move.
One clear outcome, not just the latest prompt.
Side quests become future work instead of disappearing.
Each run records what changed and what was verified.
The next agent starts from a safe, runnable task.
Why Linear and Jira are not enough
Human-updated status fields drift behind the actual work.
The graph is reconciled from runs, deltas, evidence, blockers, and decisions.
Important reasoning sits in comments, Slack, or a forgotten agent transcript.
Context is typed: goals, tasks, questions, branches, debt, and proof stay connected.
A human still decides what an agent should read, trust, ignore, and do next.
The graph selects the next agent-safe slice and explains why other work is blocked.
Pricing
For solo founders and small teams using coding agents on real product work. Continuity gives every run graph memory, blockers, proof, and a next safe task.
Early access
For founders and small teams tired of re-briefing agents, recovering lost context, and turning every handoff into archaeology.