| name | take-task |
| description | This skill should be used when claiming a task to work on - "take task", "візьми таск", "працюю над", "work on task X", "pick task X", "switch to task X", "claim task". Sets task to in_progress, creates/switches branch, saves the per-branch task-state file. |
Take Task
Claim a tracker task and prepare the working environment.
IMPORTANT: If the spec below invokes any deferred tool (e.g. AskUserQuestion), you MUST load and call it as specified. Never skip the call, never substitute a default answer of your own — the tool invocation is part of the contract. This is critical.
Why This Exists
Multiple flows need "claim task + prepare branch": /lets:start, mid-session task switch, direct "візьми таск X". Centralizing ensures consistent branch naming, worktree handling, and session ref saving.
Flow
Step 1: Resolve Task
show task=<task-id>
Verify task exists. Then claim it (set-status is a state change - HARD-FAIL loud if the binding can't run; do NOT proceed to branch setup on a failed claim):
set-status task=<task-id> status=in_progress
Step 2: Check Uncommitted Changes
Before switching branches, check for uncommitted changes:
git status --short
If changes exist and branch switch is needed, ask user:
AskUserQuestion(
questions=[{
question: "Uncommitted changes on current branch. What to do?",
header: "Uncommitted",
options: [
{ label: "Stash", description: "git stash, switch branch, remind to pop later" },
{ label: "Commit first", description: "Commit changes, then switch branch" },
{ label: "Stay", description: "Skip branch switch - work on current branch" }
],
multiSelect: false
}]
)
Handle response:
- Stash ->
git stash, proceed with branch switch, remind to git stash pop later
- Commit first -> delegate to commit skill, then proceed with branch switch
- Stay -> skip branch switch, warn about mixed work, then continue to Step 5 (the task was already claimed in Step 1 - save its
.task state file for the current branch; do NOT stop before it, or detect-task//lets:done are left with a claim but no boundary file).
If staying on current branch (worktree, already correct) or no changes - skip this step.
Step 3: Worktree Check
GIT_DIR=$(git rev-parse --git-dir 2>/dev/null)
If in a worktree ($GIT_DIR contains worktrees/):
- Skip branch creation - use the current worktree branch as-is
- Two branch shapes possible (
/lets:worktree create supports both):
worktree-<task-id>-<slug> — new-branch mode (the LETS convention); auto-detect task ID- any other shape (e.g.
feature/foo) — attached existing branch via lets worktree create --attach; no task ID in the name
- Task-id pattern is TRACKER-DEPENDENT (see detect-task): beads
<prefix>-<alphanum>[.<number>] (e.g. lets-abc, lets-abc.1); a numeric-id tracker uses a pure-numeric id
- If task ID found in branch name: confirm with user via the tracker's
show verb
- If branch shape is "attached" (no matching task ID): rely on the task-id passed as the skill argument — don't try to extract from the branch name
- Present: "In worktree, using branch: {branch}"
- Jump to Step 5
If in main repo ($GIT_DIR is .git):
Step 4: Branch Logic (main repo only)
Branch naming: feature/<task-id>-<slugified-title>
Slug rules: lowercase, spaces to hyphens, remove special chars, max 50 chars.
Examples: feature/proj-ch15-fix-proxy-config, feature/proj-ch5-add-mobile-api
Check current state:
- Already on correct branch -> do nothing, continue to Step 5
- Branch exists elsewhere ->
git checkout <branch>
- Branch doesn't exist -> offer choice:
AskUserQuestion(
questions=[{
question: "How do you want to work on this task?",
header: "Workspace",
options: [
{ label: "Branch (Recommended)", description: "Regular feature branch in current repo" },
{ label: "Worktree", description: "Separate directory for parallel work in another terminal" },
{ label: "Stay on current branch", description: "Skip branch creation. On {LETS_MERGE_BRANCH}: /lets:done pushes + closes (no PR). On a custom branch: normal PR flow." }
],
multiSelect: false
}]
)
Handle response:
- Branch ->
git checkout -b <branch> {LETS_MERGE_BRANCH} (from LETS Config)
- Worktree -> invoke
Skill(skill: "lets:worktree", args: "create <task-id>-<slug>"), then inform:
"Worktree created. Open a new terminal and run:"
cd {absolute-worktree-path} && claude
"Then use /lets:start to pick a task."
Stop here - the worktree session continues in a separate terminal.
- Stay on current branch -> skip
git checkout -b; stay on the current branch (could be $LETS_MERGE_BRANCH or any pre-existing branch). Print one line:
"Staying on {current branch}. No new branch created."
If HEAD == $LETS_MERGE_BRANCH, append: " Trunk-mode: /lets:done will push + close (no PR — same-source-target)."
Continue to Step 5 (the .task state file is saved as usual).
Step 5: Save Task State File
Write the per-branch task-state file .lets/sessions/.task-<branch-slug> with three fields:
task: (identity), start: (task boundary - preserved across resume of the same task, reset on a new/changed task or when the recorded SHA is no longer an ancestor of HEAD), and session: <sha> <session-id> (this session's boundary, always refreshed). Atomic (tmp+mv) so a crash never leaves a half-written file. <task-id> is the claimed id; {LETS_MERGE_BRANCH} is from LETS Config.
LETS_PROJECT_ROOT=$(git rev-parse --show-toplevel)
BRANCH=$(git branch --show-current); BRANCH_SLUG=$(echo "$BRANCH" | tr '/' '-')
mkdir -p "$LETS_PROJECT_ROOT/.lets/sessions"
TASK_FILE="$LETS_PROJECT_ROOT/.lets/sessions/.task-${BRANCH_SLUG}"
HEAD_SHA=$(git rev-parse HEAD)
MERGE_BRANCH="{LETS_MERGE_BRANCH}"; CLAIMED_ID="<task-id>"; SID="$CLAUDE_CODE_SESSION_ID"
PREV_TASK=""; PREV_START=""
if [ -f "$TASK_FILE" ]; then
PREV_TASK=$(sed -n 's/^task: //p' "$TASK_FILE" | head -1)
PREV_START=$(sed -n 's/^start: //p' "$TASK_FILE" | head -1)
fi
if [ "$PREV_TASK" = "$CLAIMED_ID" ] && [ -n "$PREV_START" ] && git merge-base --is-ancestor "$PREV_START" HEAD 2>/dev/null; then
START="$PREV_START"
else
START="$HEAD_SHA"
fi
tmp=$(mktemp "${TASK_FILE}.XXXX")
{
echo "task: $CLAIMED_ID"
echo "start: $START"
echo "session: $HEAD_SHA $SID"
} > "$tmp" && mv -f "$tmp" "$TASK_FILE"
Step 6: Context Recovery (existing branch)
If the branch already existed (continuing a multi-session task):
show task=<task-id>
comment-list task=<task-id>
Read the full description and ALL comments — they hold the multi-session context. Present: "Resuming {task title} ({task-id}). Last session: {summary from the latest tracker comment}"
Also surface the most recent session snapshot for this branch (the file trail /lets:start full-mode reads - the quick <task-id> path skips Step 1, so read it here):
LETS_PROJECT_ROOT=$(git rev-parse --show-toplevel)
BRANCH_SLUG=$(git branch --show-current | tr '/' '-')
LATEST_SNAP=$(ls -t "$LETS_PROJECT_ROOT/.lets/sessions/"*"-${BRANCH_SLUG}.md" 2>/dev/null | head -1)
If $LATEST_SNAP is non-empty, Read it and fold its ### Remaining + NEXT STEP into the "Last session: ..." line. If empty, rely on the tracker's comments alone (fresh branch - no snapshot yet).
Output
When triggered standalone (not via /lets:start):
Task claimed: **{task title}** (`{task-id}`)
Branch: {branch-name}
┌─ LETS ────────────────────────────────────────────┐
│ Plan? /lets:plan · --fast · plan-workflow │
│ Check? /lets:check │
└───────────────────────────────────────────────────┘
When invoked by /lets:start OR /lets:done (via Skill tool) - skip this output, the calling command has its own. Per CLAUDE.md "Internal invocation = no box".
Anti-patterns
- Never create feature/ branch inside a worktree
- Never work without saving session start ref
- Never skip the
set-status task=<id> status=in_progress claim
- Never auto-create a feature branch when user picked "Stay on current branch"
Integration
User-facing skill. Auto-triggers on "take task X", "візьми таск". Primary consumer: /lets:start.
See: grep -r "take-task" commands/ for command references.
