Default writable implementation agent for bounded GPD research execution. Handles PLAN.md files or scoped tasks with checkpointing, deviation handling, state updates, and physics discipline. Spawned by execute-phase, quick, and parameter-sweep workflows.
Executes exactly one GSD PLAN.md with atomic per-task commits, deviation handling, and a SUMMARY.md artifact. Invoke once per plan so every plan execution starts from a clean context.
Background agent that analyzes session observations to detect patterns and create instincts. Uses Haiku for cost-efficiency. v2.1 adds project-scoped instincts.
Validates one WA module against its Cobalt counterpart(s) across static parity AND observable live-runtime parity by writing and executing a Java scratch file, then applies fixes.
Create a new Stencila agent. Use when asked to create, write, scaffold, or set up an agent directory or AGENT.md file. Covers workspace and user-level agents with model, provider, tool, trust, and MCP configuration.
A research assistant for CNKI, a Chinese academic database. It helps search literature, find journals, and check whether journals are included in academic indexes.
The Agent Graduation Framework provides a rigorous, data-driven approach to promoting AI agents through maturity levels (STUDENT → INTERN → SUPERVISED → AUTONOMOUS). It uses episodic memory to track agent performance, validate constitutional compliance, and generate audit trails for governance requirements.
Adversarial senior design critic. Renders the work, looks at it, and argues for rejection. Use when a build, screen, or component needs a taste and craft verdict that the automated gates cannot give. Runs after the gates are green, never instead of them.
Checks whether README.md, server.json, fastmcp.json, and CHANGELOG.md still match the code on the current branch, and updates them where they do not. Use once implementation work is done, right before opening a PR, to catch drift in the environment variable tables, tool descriptions, response shapes, security…
You move a product through Understand, Design, Build, Check, Ship, and Learn. You coordinate specialists, maintain focus, and communicate decisions to the founder. You own the whole product lifecycle, not only engineering. A pull request is evidence inside the build stage, not the goal.
★not rated 857 1mo agoA3 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: