Skip to content

Day-to-Day Workflow

Once your project is connected and initially synced, the recurring loop is:

1. abc sync                     — pull the latest artifacts from the warehouse
2. code with agent              — agent uses synced contexts, knowledge, and skills
3. abc warehouse status         — see what changed in the warehouse working tree
4. abc warehouse contribute     — commit improvements back to the warehouse
5. repeat

Step 1: Pull Warehouse Updates

When a teammate adds a new context or improves a skill, pull and re-sync:

cd ~/team-warehouse && git pull
cd ~/projects/my-service && abc sync

✓ Sync complete
  Updated: 1 symlink   (skills/code-review/SKILL.md)
  Created: 0 symlinks
  Up to date: 11 symlinks

If the warehouse has new artifacts you haven't adopted yet:

✓ Sync complete
  Up to date: 11 symlinks

2 new artifact(s) available — run abc adopt to review

Run abc adopt to open the TUI and selectively add them.

Step 2: Code with Your Agent

Your agent reads at session start:

  • Contexts wired into CLAUDE.md / opencode.json — loaded automatically
  • Knowledge — markdown files symlinked into artifacts/
  • Skills — available as slash commands (/code-review, /generate-tests, etc.)

No extra setup needed — everything is in place after abc sync.

Example: agent improves a knowledge file mid-session

You ask the agent to document a new error handling pattern. It writes the explanation directly into artifacts/knowledge/python/error-handling.md. Because that file is a symlink into your warehouse clone, the warehouse working tree is updated immediately.

No copy step needed — the improvement is already in the warehouse.

Step 3: Review Warehouse Changes

After a coding session, check what the agent touched:

abc warehouse status

Modified files in warehouse:
  modified  knowledge/python/error-handling.md
  modified  skills/code-review/SKILL.md

Inspect a specific file before committing:

abc warehouse status knowledge/python/error-handling.md

--- a/knowledge/python/error-handling.md
+++ b/knowledge/python/error-handling.md
@@ -12,6 +12,14 @@
 ## Targeted try/except

 Keep `try` blocks small and targeted — one concern per block.
+
+## Retrying with tenacity
+
+Use `@retry` from tenacity instead of manual sleep loops:
+
+```python
+@retry(stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=1, max=10))
+def fetch_data(url: str) -> dict: ...
+```

Step 4: Contribute Changes Back

Contribute everything at once

abc warehouse contribute -m "docs: add tenacity retry pattern to error handling"

This stages all modified warehouse files tracked by beacon.yaml, commits them, and exits.

Contribute and push immediately

abc warehouse contribute -m "docs: add tenacity retry pattern" --push

Teammates get the update on their next git pull && abc sync.

Contribute only specific files

Use plain git inside the warehouse clone to stage selectively:

cd ~/team-warehouse
git add knowledge/python/error-handling.md
git commit -m "docs: add tenacity retry pattern"
git push

Leave skills/code-review/SKILL.md unstaged if you're not ready to share that change yet.

Discard a change you don't want to keep

A common case: you ran abc warehouse status and find two modified files, but only one is the change you intended. The other is a stray edit from earlier in the session — perhaps the agent retitled a skill in passing. Discard it from the warehouse clone:

cd ~/team-warehouse
git checkout -- skills/code-review/SKILL.md

The symlink reflects the restored file immediately in your project — no further sync needed.

Checking Project Health

abc status      # connected warehouse, synced artifacts, ✓/✗ per item
abc doctor      # full validation; --fix to auto-resolve stale paths

Next Steps