Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
Goal Forge runs a rough idea through a structured pipeline—interview, tighten, compile, config check—to produce a Codex /goal contract. It forces explicit scorecards, feedback loops, working memory, and measurable done_when criteria before any autonomous execution begins. This skill gives AI agent engineers, platform teams, and technical leads a repeatable way to move from vague intent to verifiable autonomous work.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Use the goal-forge skill in Interview mode. Interview me about this rough coding idea: [describe the idea]. Ask decision-forcing questions about relevant scope, architecture, edge cases, tradeoffs, verification, risks, and non-goals. Once the idea is complete enough to compile, write or update `SPEC.md`.
Any constraints or decisions that must remain unchanged
Use the goal-forge skill in Tighten mode on the existing specification at [path/to/SPEC.md]. Challenge unclear decisions and help make its scope, scoring, feedback, and verification concrete enough for execution.
Use the goal-forge skill in Check config mode from the root of [path/to/target-project]. Produce the documented read-only readiness report for the Codex version, project trust, and autonomous `/goal` configuration; do not change the configuration.
Codex