Skip to content

Docs/agent workflow rules#30

Merged
nah414 merged 4 commits into
mainfrom
docs/agent-workflow-rules
Jun 9, 2026
Merged

Docs/agent workflow rules#30
nah414 merged 4 commits into
mainfrom
docs/agent-workflow-rules

Conversation

@nah414

@nah414 nah414 commented Jun 9, 2026

Copy link
Copy Markdown
Owner

Summary by Sourcery

Document repository-specific operating rules and workflow for AI agents working in the project, and surface them prominently in the docs.

Documentation:

  • Add AGENT_WORKFLOW.md with detailed agent operating rules, halt conditions, branching/PR conventions, and reporting expectations.
  • Introduce CLAUDE.md at the repo root as the short-form agent rules that auto-load in Claude Code sessions, pointing to the detailed workflow docs.
  • Update docs/README.md to reference the new agent workflow documentation alongside other internal development records.

nah414 and others added 4 commits June 9, 2026 14:14
docs(agents): add CLAUDE.md auto-loaded agent operating rules

Encode the maintainer's working rules so every Claude Code session in
this repo picks them up automatically (CLAUDE.md auto-loads per-repo).

- CLAUDE.md (root): concise rule set, points to the detailed doc.
- docs/AGENT_WORKFLOW.md: long-form rules, halt list, branch/commit
  conventions, reporting format, Windows/PowerShell notes, rationale.
- docs/README.md: link the new workflow doc from the docs hub.

Rules optimize for remote phone-driven sessions via Dispatch: no
interactive pickers (plain-text questions only), autonomous reversible
defaults, halt only for destructive/irreversible actions, feature
branch only, plan-but-do-not-block, test before handing back.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@
docs(agents): forbid skipping git hooks / signing as a top-level rule

Add an explicit short-form rule (new rule 5) in CLAUDE.md: never use
--no-verify, never disable/skip/bypass pre-commit or commit-msg hooks or
commit signing unless the maintainer explicitly asks; fix the underlying
issue instead. Mirrors and strengthens AGENT_WORKFLOW.md section 4.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@
docs(agents): post-reorg path fixes + PR-consolidation rule

Rebased onto current main (post #26/#27 docs reorg). Adjust for the new
docs/ tree and capture the maintainer's PR-cadence preference.

- CLAUDE.md: repoint build-guide reference from the old repo-root
  BUILDGUIDE_*.md files to docs/build-guides/ (moved by #27).
- docs/README.md: re-add the AGENT_WORKFLOW.md pointer into the
  reorganized "Internal development records" section (the old pre-reorg
  edit was dropped during rebase).
- CLAUDE.md + AGENT_WORKFLOW.md: add a new rule (short form rule 5;
  detail in AGENT_WORKFLOW.md section 4) — consolidate work, at most one
  PR per work-session/batch, branches may accumulate commits and sit for
  days, never open a PR proactively, default posture is keep-on-branch.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@
docs(agents): record mobile-Dispatch working method + 2026-06-09 session log

Note in CLAUDE.md (not the README) that this repo is primarily driven via
mobile Dispatch (Adam running Claude Code sessions from his phone while the
desktop runs), and log today's session: README trim (#26 merged), docs/
reorg (#27 merged), auto-capture wiring (phase-13.x.9, parked), these agent
rules, and the Step 5c cognition training/eval harness (parked; lightgbm
4.6.0 kept for local runs).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@
@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

@sourcery-ai

sourcery-ai Bot commented Jun 9, 2026

Copy link
Copy Markdown

Reviewer's Guide

Adds explicit operating rules for AI agents working in this repo, via a new long-form workflow doc under docs/ and a concise auto-loaded CLAUDE.md at the root, and links the new workflow doc from the docs README.

Flow diagram for CLAUDE.md and AGENT_WORKFLOW.md documentation linkage

flowchart TD
    AgentSession[Agent session starts in repo]
    CLAUDE["CLAUDE.md (short operating rules)"]
    AGENT["docs/AGENT_WORKFLOW.md (long-form workflow rules)"]
    DocsREADME["docs/README.md"]

    AgentSession --> CLAUDE
    CLAUDE --> AGENT
    DocsREADME --> AGENT
Loading

File-Level Changes

Change Details Files
Document detailed operating rules and conventions for AI agents working in the repository.
  • Create docs/AGENT_WORKFLOW.md with long-form guidance on agent behavior, halt-list actions, branch/commit conventions, PR cadence, planning expectations, test requirements, reporting format, and Windows/PowerShell specifics.
  • Explain rationale emphasizing remote, phone-driven supervision and the need for reversible, autonomous progress.
docs/AGENT_WORKFLOW.md
Introduce a short-form, auto-loaded CLAUDE.md summarizing the key agent rules and repo-specific context.
  • Add CLAUDE.md at the repo root with a concise numbered list of operating rules referencing the detailed workflow doc.
  • Document Phoenix-specific context (architecture spec, phase-gated development, platform expectations) and note the primary mobile/Dispatch workflow.
  • Add an initial dated session log capturing recent structural and feature work plus these agent rules.
CLAUDE.md
Update documentation index to reference the new agent workflow doc.
  • Add AGENT_WORKFLOW.md entry under internal development records in docs/README.md, describing it as the long-form companion to CLAUDE.md.
docs/README.md

Tips and commands

Interacting with Sourcery

  • Trigger a new review: Comment @sourcery-ai review on the pull request.
  • Continue discussions: Reply directly to Sourcery's review comments.
  • Generate a GitHub issue from a review comment: Ask Sourcery to create an
    issue from a review comment by replying to it. You can also reply to a
    review comment with @sourcery-ai issue to create an issue from it.
  • Generate a pull request title: Write @sourcery-ai anywhere in the pull
    request title to generate a title at any time. You can also comment
    @sourcery-ai title on the pull request to (re-)generate the title at any time.
  • Generate a pull request summary: Write @sourcery-ai summary anywhere in
    the pull request body to generate a PR summary at any time exactly where you
    want it. You can also comment @sourcery-ai summary on the pull request to
    (re-)generate the summary at any time.
  • Generate reviewer's guide: Comment @sourcery-ai guide on the pull
    request to (re-)generate the reviewer's guide at any time.
  • Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
    pull request to resolve all Sourcery comments. Useful if you've already
    addressed all the comments and don't want to see them anymore.
  • Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
    request to dismiss all existing Sourcery reviews. Especially useful if you
    want to start fresh with a new review - don't forget to comment
    @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

  • Enable or disable review features such as the Sourcery-generated pull request
    summary, the reviewer's guide, and others.
  • Change the review language.
  • Add, remove or edit custom review instructions.
  • Adjust other review settings.

Getting Help

@sourcery-ai sourcery-ai Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hey - I've reviewed your changes and they look great!


Sourcery is free for open source - if you like our reviews please consider sharing them ✨
Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.

@nah414 nah414 merged commit a94fed8 into main Jun 9, 2026
19 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant