Implement Plan

Implement Plan

Execute an implementation plan file.

Task Tracking

At the start, use update_plan to track each step:

1. Resolve and read the plan file
2. Read context files
3. Load task-specific skills
4. Add plan Implementation Steps to the active plan
5. Run $implement skill
6. Update plan status

Step 1: Resolve and Read the Plan File

Determine which plan file to implement using these rules in order:

1. Explicit path — If an absolute or relative path was passed, use it
2. Explicit slug — If a slug was passed (e.g., add-image-cache), resolve to .turbo/plans/<slug>.md if that file exists; otherwise fall through to rule 5 before erroring
3. Single file — Glob .turbo/plans/*.md. If exactly one plan exists, use it
4. Most recent — If multiple plans exist, use the most recently modified
5. Unexpanded shell — If a slug or path was passed but resolved to no plan, derive <slug> from the basename (stripping .md) and check .turbo/shells/<slug>.md. If it exists, halt with: "<slug> is a shell that needs expansion first. Run $pick-next-shell to expand and implement it."
6. Legacy fallback — If .turbo/plans/ does not exist but .turbo/plan.md exists, use it
7. Nothing found — If no plan matched any rule above, tell the user to run $turboplan (for a new task) or $pick-next-shell (for existing shells) and stop

If multiple plans exist and the most-recent choice is non-obvious (e.g., several plans were modified within the same minute), use request_user_input to let the user pick from the candidates.

State the resolved plan path before continuing, then read the file.

Step 2: Read Context Files

Read in full:

• Every file listed in the plan's Context Files section
• Files the user referenced in their original request (if any)
• Every file path the plan references in the Context, Pattern Survey, and Implementation Steps sections

Step 3: Load Task-Specific Skills

Scan the plan's Implementation Steps for work types that match available skills. For each unambiguous match, run the skill by reading and following the installed skill instructions. For example, if the plan includes "add a Drizzle migration" and a skill exists whose triggers reference Drizzle migrations, load it. If the plan mentions "run the test suite" but no testing-specific skill trigger matches, do not load a generic testing skill.

If unsure, do not load. Do not load $code-style here.

Step 4: Add Plan Implementation Steps to the Active Plan

Append one plan entry per plan Implementation Step. These entries belong to $implement; do not mark them in_progress or completed in this skill. Codex update_plan is flat — there is no parent/child nesting; use clear step text so each entry is self-explanatory in a flat list.

Step 5: Run $implement Skill

Run the $implement skill. The plan file, its file references, and its Verification section are already in conversation context from Step 1.

Step 6: Update Plan Status

After $implement completes, set the plan's frontmatter status: to done. If the plan is the legacy .turbo/plan.md without frontmatter, skip this step.

Rules

• The plan file is read-only during execution. If revisions are needed, run $refine-plan or $draft-plan separately.
• Never skip Steps 2 or 3.
• Never enumerate or execute the plan's Implementation Steps inline. The work runs through $implement. Restating steps as a turn-level narration counts as inline execution and bypasses the delegation.
• If the plan's Implementation Steps or Verification include git commit, git push, or PR creation, halt before Step 5 and ask the user to remove them via $refine-plan.