skills.rest
AgentWorkforce
AgentWorkforceOFFICIAL
Explore all skills by this author →
Website𝕏@agent_relayGitHub

using-trajectories-sdk

Capture and persist agent work trajectories with TrajectoryClient and TrajectoryBuilder.

28|4|Updated Dec 20, 2025
Search Tags:#automation#typescript#sdk#agent#persistence#trajectory#trajectories
One-click install
npx skills add https://github.com/AgentWorkforce/trajectories --skill using-trajectories-sdk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: using-trajectories-sdk
Source: https://github.com/AgentWorkforce/trajectories/tree/main/.openskills/using-trajectories-sdk
Command: npx skills add https://github.com/AgentWorkforce/trajectories --skill using-trajectories-sdk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Programmatically capture and persist agent work trajectories, enabling structured, searchable records of how tasks were approached, decisions made, and results achieved.

Core Features & Use Cases

  • TrajectoryClient (persistent storage) for saving trajectories to disk and sharing across tools.
  • TrajectoryBuilder (in-memory construction) for quickly composing trajectories without persistence.
  • Exports to Markdown, JSON, and timeline formats for integration with tooling.
  • Common use cases: building automation that records exploration, decisions, and retrospectives; CI/CD pipelines that capture task context alongside code changes.

Quick Start

Install the agent-trajectories package, initialize a TrajectoryClient, and start a trajectory to begin recording work.

Frequently Asked Questions about using-trajectories-sdk

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I programmatically capture and persist agent work trajectories in TypeScript?▼

You can capture and persist agent work trajectories using a TrajectoryClient to save structured, versioned records of task exploration and decisions directly to disk for consistent retrieval across tools.

How do I record agent decisions in a CI/CD pipeline?▼

Recording agent decisions in a CI/CD pipeline involves initializing a trajectory to capture task context alongside code changes, ensuring versioned records of exploration, decisions, and outcomes are stored consistently.

Can I build agent trajectories in memory without persisting them to disk?▼

Yes, you can construct agent trajectories in memory without persistence by using a TrajectoryBuilder, allowing you to quickly compose work records before deciding whether to export or save them.

What export formats are supported for agent trajectory records?▼

Agent trajectory records support exports to Markdown, JSON, and timeline formats, enabling seamless integration with various tooling and automation workflows for structured data analysis.

What is the best way to share agent task context across multiple automation tools?▼

The best way to share agent task context across tools is by using persistent storage via TrajectoryClient, which saves structured trajectory records to disk for reliable access by different automation components.

Related Skills

Explore More Software Engineering Skills→
sandmark78sandmark78

agent-optimizer

Record execution trajectories and gather rewards to optimize agent performance.

Community
Intermediate
hairyfhairyf
4

aderyn

Analyze Solidity smart contracts for vulnerabilities and code quality issues.

Community
Intermediate
4lerman4lerman

ci-cd-pipeline-builder

Detect repository stacks and generate CI/CD pipeline templates for GitHub Actions and GitLab CI.

Community
Advanced
ksopylaksopyla

engineering-tracker

Automate CHANGELOG.md updates with engineering change records.

Community
Intermediate