Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/taoidle/plan-cascade/completegit clone --depth 1 https://github.com/Taoidle/plan-cascadeWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00051 | $0.03435 |
| Opus 5 | $0.00026 | $0.01717 |
| Sonnet 5 | $0.00010 | $0.00687 |
| Haiku 4.5 | $0.00005 | $0.00344 |
Grade C, and why
complete scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf ".state" How it starts
The opening of the file, as written. The whole thing — 461 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Planning with Files - Complete Worktree Task
You are now completing a worktree task. This will:
- Verify all phases are complete
- CRITICAL: Commit code changes (planning files excluded)
- Delete planning files from the worktree
- Navigate to the root directory
- Merge the task branch to target branch
- Remove the worktree
- Delete the task branch
Path Storage Modes
This command works with both new and legacy path storage modes:
New Mode (Default)
- Worktrees in:
~/.plan-cascade/<project-id>/.worktree/ - State files in:
~/.plan-cascade/<project-id>/.state/ - Cleanup removes files from user data directory
Legacy Mode
- Worktrees in:
<project-root>/.worktree/ - State files in project root
The command auto-detects which mode is active based on .planning-config.json contents.
Step 1: Detect Current Location
Check if we're in a worktree or the root directory:
if [ -f ".planning-config.json" ]; then
# We're in a worktree directory
echo "Currently in worktree directory: $(pwd)"
IN_WORKTREE=true
else
# We're not in a worktree, check if there are any worktrees
IN_WORKTREE=false
# Check for worktrees
WORKTREES=$(git worktree list 2>/dev/null | grep -v "\bare$" | wc -l)
if [ "$WORKTREES" -eq 0 ]; then
echo "ERROR: No worktrees found."
echo "This command requires an existing worktree."
echo ""
echo "Create one first with:"
echo " /plan-cascade:worktree <task-name> <branch>"
exit 1
fi
echo "Not in a worktree directory. Found $WORKTREES worktree(s):"
echo ""
git worktree list
echo ""
# Find all worktrees with .planning-config.json (check both new and legacy locations)
echo "Scanning for planning worktrees..."
WORKTREE_LIST=()
# Get worktree base directory from PathResolver
WORKTREE_BASE=$(uv run python -c "from plan_cascade.state.path_resolver import PathResolver; from pathlib import Path; print(PathResolver(Path.cwd()).get_worktree_dir())" 2>/dev/null || echo ".worktree")
while IFS= read -r line; do
worktree_path=$(echo "$line" | awk '{print $1}')
worktree_branch=$(echo "$line" | awk '{print $2}')
# Check if this is a planning worktree
if [ -f "$worktree_path/.planning-config.json" ]; then
# Exclude hybrid mode worktrees (those use /plan-cascade:hybrid-complete)
mode=$(jq -r '.mode // empty' "$worktree_path/.planning-config.json" 2>/dev/null)
if [ "$mode" != "hybrid" ]; then
task_name=$(jq -r '.task_name // empty' "$worktree_path/.planning-config.json" 2>/dev/null)
WORKTREE_LIST+=("$worktree_path|$task_name|$worktree_branch")
fi
fi
done < <(git worktree list 2>/dev/null | grep -v "\bare$")
if [ ${#WORKTREE_LIST[@]} -eq 0 ]; then
echo "ERROR: No planning worktrees found."
echo "Found worktrees but none are in planning mode."
echo ""
echo "Note: Hybrid mode worktrees should use /plan-cascade:hybrid-complete"
exit 1
fi
echo "Found ${#WORKTREE_LIST[@]} planning worktree(s):"
echo ""
# Display options
for i in "${!WORKTREE_LIST[@]}"; do
IFS='|' read -r path name branch <<< "${WORKTREE_LIST[$i]}"
echo " [$((i+1))] $name"
echo " Path: $path"
echo " Branch: $branch"
echo ""
done
# Ask user to select
echo "Which worktree would you like to complete?"
read -p "Enter number (or 0 to cancel): " selection
if [ "$selection" = "0" ]; then
echo "Cancelled."
exit 0
fi
if [ "$selection" -lt 1 ] || [ "$selection" -gt ${#WORKTREE_LIST[@]} ]; then
echo "Invalid selection."
exit 1
fi
# Get the selected worktree
selected="${WORKTREE_LIST[$((selection-1))]}"
IFS='|' read -r WORKTREE_PATH TASK_NAME TASK_BRANCH <<< "$selected"
echo ""
echo "Selected: $TASK_NAME"
echo "Navigating to worktree: $WORKTREE_PATH"
# Change to worktree directory
cd "$WORKTREE_PATH" || {
echo "ERROR: Failed to navigate to worktree: $WORKTREE_PATH"
exit 1
}
echo "✓ Now in worktree: $(pwd)"
IN_WORKTREE=true
fi
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 461 lines · 51 tokens per session scan C 68488663ef33
complete is a command published in the GitHub repository Taoidle/plan-cascade (124 stars, last pushed 5mo ago), licensed MIT. It adds 51 tokens to every session and 3,435 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
conv_release
Command "conv_release" from John-Wendell/Attention-MoA, covering chatbot arena conversations, all conversations and prompt distribution.
data_cleaning
Command "data_cleaning" from John-Wendell/Attention-MoA, covering data cleaning, requirements, steps, convert html to markdown and keep or remove specific languages.
pypi
Command "pypi" from John-Wendell/Attention-MoA, covering requirement and upload.
local_cluster
node-01.
webserver
Command "webserver" from John-Wendell/Attention-MoA, covering install, launch servers, check the launch time, increase the limit of max open files and gradio edit (3.35.2).
leaderboard
Command "leaderboard" from John-Wendell/Attention-MoA, covering get logs, clean battle data, run elo analysis, copy files to hf space and update files on webserver.