Trains reusable natural-language 'skills' for frozen LLM agents by optimizing the skill document in text-space — using trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts. Multi-backend, zero inference-time cost at deployment, designed for iterative, validation-led skill improvement.
Provides long-term memory for AI coding agents by compiling sanitized lifecycle observations into a git-versioned Markdown wiki that enables cross-agent handoffs, per-project isolation, and optional vector-backed retrieval.
Moves repository search into a dedicated exploration subagent that issues parallel read-only READ/GLOB/GREP calls and returns compact file:line citations. Trained (4B–30B) with SFT+RL, it reduces main-agent token use up to ~60% and raises end-to-end success by up to ~5.5%.
Captures an app's network traffic, extracts its authentication/session bundle, and auto-generates an editable Python client to call the app's API. Supports mitmproxy, cURL and HAR capture backends and provides Session helpers for replay and token refresh.
Provides a unified tool catalog and proxy for AI agents: one team token lets agents call thousands of third-party endpoints (and your team’s keys) with per-call pricing. Injects credentials server‑side so agents never hold provider secrets, and exposes CLI/skill recipes and OAuth tools for team-wide use.
Rewrites agent-facing English into ASD-STE100 Simplified Technical English to remove parsing ambiguity for downstream AI agents. Performs sentence-by-sentence rule checks and deterministic rewrites while preserving all facts and scope qualifiers. Best for error messages, tool descriptions, and inter-agent instructions.
Evaluates whether coding agents can modify real scientific software while preserving domain-specific scientific contracts. Contains 119 repository-level tasks across 98 GitHub projects and 20 scientific domains, measures reproducible edits in pinned Docker images, and analyzes recurring failure modes.