ToruAI

Map unfamiliar code or research an approach before planning, read-only and fast.

Run the project's own tests, linter, type checker and build, and report what actually failed.

The last gate before merge — QA, security audit, review and changelog, with objective pass conditions.

Look up how Claude Code itself works — hooks, skills, subagents, MCP, plugins, settings — from the official docs, so answers come from the current spec instead of memory.

Run a feature end to end — scout the code, plan it, build it, and keep a session file so a closed terminal doesn't lose the thread.

Close out a session — summarise what changed, capture what was learned, and prepare the commit.

Save a project fact, decision or gotcha so your next session already knows it.

Hand off an unfinished session to the next one — what's done, what's next, what's blocked.

Audit the code for vulnerabilities, leaked secrets and risky dependencies before it ships.

Hand over a spec and walk away — plans, builds, tests and audits on its own, stopping only at a real blocker.

Branch and commit conventions that keep an agent's work reviewable — one branch per task, main stays clean, nothing force-pushed.

Turn friction you just hit into a concrete change to the workflow, so the same problem doesn't recur.

Write the spec before the code, then check the built thing against it — so agents build what was agreed, not what they invented.

Design better agents, skills and hooks for Claude Code, checking the current spec instead of guessing at fields.