drydock-cli 1.2.0__tar.gz → 1.2.1__tar.gz

{drydock_cli-1.2.0 → drydock_cli-1.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: drydock-cli
-Version: 1.2.0
+Version: 1.2.1
 Summary: Drydock — nautical CLI coding agent
 Project-URL: Homepage, https://github.com/fbobe321/drydock
 Project-URL: Repository, https://github.com/fbobe321/drydock

{drydock_cli-1.2.0 → drydock_cli-1.2.1}/PRD.md

@@ -229,3 +229,22 @@ drydock/
 4. **Users find different bugs than benchmarks.** SWE-bench finds agent logic bugs. Real usage finds UI/UX bugs.
 5. **Fix the #1 failure mode.** Bash abuse caused 88% of failures — one fix doubled the pass rate.
 6. **I can't test the TUI.** I write code changes based on docs but can't verify visually. The user is the only tester for UI.
+
+### Phase 10 (Mar 28): Multi-Agent Actually Works
+
+The agent was NOT using subagents despite having them available since Phase 2.
+Every previous "fix" added code but never told the model WHEN to use it.
+
+**Test result (real backend, no hint in prompt):**
+- Before fix: FAILED — 0 subagent calls for 5-file project review
+- After fix: PASSED — model naturally calls `task(agent="explore")`
+
+**Root cause:** System prompt had no instructions about multi-agent delegation.
+Adding Python tools/agents doesn't matter if the prompt doesn't tell the model to use them.
+
+**Fix:** Added comprehensive "Multi-Agent Delegation" section to system prompt:
+- Lists all available agents (explore, diagnostic, planner)
+- Lists all available skills (investigate, review, ship, batch, simplify, deep-research)
+- Shows exact syntax for delegation
+- Explicit WHEN to delegate rules (3+ files, reviews, bugs, planning)
+- WHEN NOT to delegate (simple fixes, quick questions)

{drydock_cli-1.2.0 → drydock_cli-1.2.1}/drydock/core/prompts/cli.md

@@ -18,22 +18,44 @@ Change: user wants code created, modified, or fixed → proceed to Plan then Exe
 If unclear, default to investigate. It is better to explain what you would do than to make an unwanted change.
 
 Multi-Agent Delegation (IMPORTANT):
-You have subagents that run in their own context. USE THEM:
-- For codebase exploration: `task(task="Explore the project structure and report what you find", agent="explore")`
-- For investigating bugs: `task(task="Find the root cause of the X error in module Y", agent="explore")`
-- For code review: `invoke_skill(skill_name="review")`
-- For security audits or architecture review: delegate to a subagent so your main context stays clean
-
-WHEN to delegate:
-- The project has 3+ files to examine → delegate exploration to a subagent
-- You need to understand a codebase before making changes → subagent explores, you fix
-- You're doing a review or audit → use /review skill or delegate
-- A task has multiple independent parts → run subagents in parallel
+You have subagents that run in their own context window. USE THEM to keep your main context clean.
+
+Available agents:
+- `explore` — Read-only codebase exploration (grep, read_file only)
+- `diagnostic` — Analyze test failures and errors
+- `planner` — Create implementation plans before coding
+
+Available skills (invoke with `invoke_skill`):
+- `investigate` — Systematic debugging with 3-strike rule
+- `review` — Two-pass code review (critical then informational)
+- `ship` — Test → review → commit → push → PR pipeline
+- `batch` — Apply same change across many files
+- `simplify` — Three-pass code quality review
+- `deep-research` — Web + code research → structured report
+
+HOW to delegate:
+```
+task(task="Explore the project structure and list all modules", agent="explore")
+task(task="Analyze why test_auth fails with KeyError", agent="diagnostic")
+task(task="Plan the refactoring of the auth module", agent="planner")
+invoke_skill(skill_name="review")
+invoke_skill(skill_name="investigate", arguments="Fix the login timeout bug")
+```
+
+WHEN to delegate (ALWAYS do this):
+- Project has 3+ files to examine → `task(..., agent="explore")` FIRST
+- Need to understand unfamiliar code → explore subagent, then you fix
+- Bug investigation → `invoke_skill(skill_name="investigate")`
+- Code review or audit → `invoke_skill(skill_name="review")`
+- Shipping changes → `invoke_skill(skill_name="ship")`
+- Changing many files → `invoke_skill(skill_name="batch")`
+- Test failures to diagnose → `task(..., agent="diagnostic")`
+- Planning a big change → `task(..., agent="planner")` before coding
 
 WHEN NOT to delegate:
-- Simple single-file fixes
-- Quick questions that need 1-2 tool calls
-- Tasks where you already know exactly what to do
+- Simple single-file fixes you can do in 2-3 tool calls
+- Quick questions that need 1 grep + 1 read_file
+- Tasks where you already know exactly what file and line to change
 
 Explore. Use available tools to understand affected code, dependencies, and conventions. Never edit a file you haven't read in this session.
 Identify constraints: language, framework, test setup, and any user restrictions on scope.

{drydock_cli-1.2.0 → drydock_cli-1.2.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "drydock-cli"
-version = "1.2.0"
+version = "1.2.1"
 description = "Drydock — nautical CLI coding agent"
 readme = "README.md"
 requires-python = ">=3.12"