Agent on Alwaldend

Slop without a clear goal

Sat, 12 Sep 2026 00:00:00 +0000

A while back, I said to Jipity¹: “Jipity, how can I improve agent ergonomics in this monorepo?”, and Jipity gave me some useful suggestions, but one of them was not like the rest. It said that context in the repo was too splintered — agents had to read many things in different directories to understand how things are done. In my infinite wisdom, I said, “Jipity, write me a comprehensive plan implementing a unified agent system with the goal of improving ergonomics in this monorepo”. And it wrote me a plan, I did not really read it.

Agents

Mon, 01 Jan 0001 00:00:00 +0000

Repository agent skills

This project owns the reusable cross-repository agent skills. Maintained improvement work lives in this project’s OpenSpec workspace. It does not centralize component facts or runtime state: each fact remains canonical at its natural owner.

Start here

Document	Purpose
OpenSpec changes	Maintained work, acceptance, and preserved goal history
Root agent guide	Current repository-wide operating policy

Dated audits remain in migrated history.

Ownership boundaries

openspec/ owns agent-system requirements, changes, and evidence; the migration map locates historical records.
Other components own their specifications and changes under their own openspec/ directory. infra/src/openspec/ describes repository evolution.
skills/ owns reusable cross-product, repository-wide agent procedures and their development-time evaluations. A product-specific procedure belongs with its product; a procedure that no narrower project genuinely owns lives here rather than in a contrived local home.
Product-specific skills remain with their project at projects/<project>/skills/<name>. Do not split one skill across owners.

Skill packaging and discovery

Every canonical skill is packaged as a skill_library in its owning directory.

Agents

Mon, 01 Jan 0001 00:00:00 +0000

This project owns the reusable, cross-project agent skills used across this repository. Each skill is a procedure with its own scope and ownership rules, so an agent loads the guidance it needs for the task at hand instead of one large document.

Features

One skill per procedure, with an explicit description and scope
Discovery written into the agent skills directory from declared sources
Product-specific skills stay with their product; only genuinely shared ones live here

Bazel agent

Mon, 01 Jan 0001 00:00:00 +0000

bazel_agent is the repository Bazel entry point for agents. It keeps the Bazel invocation behind a validated subcommand while consistently applying the agent configuration:

bazel_agent bazel test //path/to/package:all
    ->
bazel test --config=agent //path/to/package:all

bazel is the only Bazel entry point, and it is a validated subcommand. The runner accepts a known Bazel command after the bazel keyword, rejects arbitrary leading arguments, and places --config=agent after the command. Targets, command options, and arguments after the -- separator of a bazel run invocation pass through unchanged. Later options therefore retain Bazel’s normal precedence and can override settings supplied by the agent configuration. The runner resolves bazel from PATH and replaces itself with that process, so signals and the final exit status are not mediated by another wrapper process. Uses of the persistent Bazel server follow the host Codex network policy; the host-bot profile allows loopback so the client-server connection is not blocked.

Bazel agent

Mon, 01 Jan 0001 00:00:00 +0000

bazel_agent is the Bazel entry point for agents working in this repository. It keeps Bazel behind a validated subcommand while consistently applying the agent configuration.

Features

Validated bazel subcommand instead of arbitrary arguments
Applies the agent configuration in the required position
Preserves signals and exit status by replacing itself with the Bazel process
Read-only doctor diagnostics for runner and workspace state