Planning with files

v2.26.1

Implements Manus-style file-based planning to organize and track progress on complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when aske...

36· 15k·147 current·154 all-time
byAhmad Othman Ammar Adi.@othmanadi
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
The skill is described as a Manus-style file-based planner and the provided templates, init script, check-complete script, and session-catchup script all support that purpose. Requested tools (Read, Write, Edit, Bash, Glob, Grep) and behaviors (create/read task_plan.md, findings.md, progress.md) align with the description. There are no unrelated environment variables or credentials requested.
Instruction Scope
SKILL.md and hooks instruct the agent to read and echo local planning files and to run the session-catchup.py script which scans session storage under the user's home (e.g., ~/.claude/projects, and conditionally ~/.codex/sessions). This is coherent for session catchup, but it does widen scope to reading previous conversation logs and assistant/user messages from disk (potentially exposing unrelated session content). Hooks also automatically run small shell commands (cat, head, tail, echo) which will reveal local file contents to the agent when invoked.
Install Mechanism
There is no remote install step or external download; the skill is instruction-only with included scripts and templates. No archives or external URLs are fetched by the skill itself.
Credentials
The skill requests no secrets or environment variables, but it assumes access to the user's filesystem (current project directory and ~/.claude/.codex session stores). Accessing session logs and project directories is consistent with the catchup feature but is a broader read scope than simple plan file creation and may expose prior conversation content.
Persistence & Privilege
always is false and the skill does not request permanent system-level privileges. The skill will create and modify files in the user's project directory (task_plan.md, findings.md, progress.md) which is expected for its purpose. It does not attempt to change other skills' configs.
Assessment
This skill appears to do what it says: create and maintain three markdown files and optionally run a 'session catchup' that inspects local session logs. Before installing or running it, consider the following: - Inspect the scripts (session-catchup.py, init-session.sh, check-complete.sh) yourself. They are short and readable; no network exfiltration is present, but session-catchup.py reads prior session JSONL files under your home directory (e.g., ~/.claude/projects and conditionally ~/.codex/sessions). If those session files contain sensitive data, the catchup output could surface it. - The skill will create/modify files in whatever directory you run it from. Run it in a project directory you control, not in a home or system directory you care about. - Hooks (PreToolUse/PostToolUse/Stop) run local shell commands (cat/head/tail/ps1/sh). That behavior is expected for reminders/read-before-decision logic, but review the commands and ensure your environment's CLAUDE_PLUGIN_ROOT resolves to the intended path. - If you are uncomfortable with the catchup behavior reading past sessions, avoid running the catchup script or remove/modify that part of SKILL.md before use. Overall the skill is internally consistent with its stated purpose, but review the included scripts and run them in a sandboxed/project-specific directory if you want to limit exposure of prior session data.

Like a lobster shell, security has layers — review code before you run it.

agentvk97axkze7k084r7wta2cnf8rr5800cs1agent-skillsvk97axkze7k084r7wta2cnf8rr5800cs1agentsvk97axkze7k084r7wta2cnf8rr5800cs1antigravityvk97axkze7k084r7wta2cnf8rr5800cs1c lawdvk97axkze7k084r7wta2cnf8rr5800cs1claudevk97axkze7k084r7wta2cnf8rr5800cs1claude-codevk97bhbkz1rxvacxk7m92mb55118391zeclaude-skillsvk97axkze7k084r7wta2cnf8rr5800cs1clawdbotvk97axkze7k084r7wta2cnf8rr5800cs1clawdbot-skillvk97axkze7k084r7wta2cnf8rr5800cs1clawdhubvk97axkze7k084r7wta2cnf8rr5800cs1context-engineeringvk97bhbkz1rxvacxk7m92mb55118391zecursorvk97bhbkz1rxvacxk7m92mb55118391zefactory-aivk97axkze7k084r7wta2cnf8rr5800cs1geminivk97bhbkz1rxvacxk7m92mb55118391zehooksvk97bhbkz1rxvacxk7m92mb55118391zekilo-codevk97bhbkz1rxvacxk7m92mb55118391zekilocodevk97axkze7k084r7wta2cnf8rr5800cs1latestvk97bhbkz1rxvacxk7m92mb55118391zemanusvk97bhbkz1rxvacxk7m92mb55118391zemanus-aivk97axkze7k084r7wta2cnf8rr5800cs1markdownvk97bhbkz1rxvacxk7m92mb55118391zemulti-idevk97bhbkz1rxvacxk7m92mb55118391zepersistent-memoryvk97bhbkz1rxvacxk7m92mb55118391zeplanningvk97bhbkz1rxvacxk7m92mb55118391zeproductivityvk97bhbkz1rxvacxk7m92mb55118391zeproject-managementvk97bhbkz1rxvacxk7m92mb55118391zeprompt-engineeringvk97axkze7k084r7wta2cnf8rr5800cs1reverse-engineeringvk97axkze7k084r7wta2cnf8rr5800cs1skillvk97bhbkz1rxvacxk7m92mb55118391zetask-planningvk97bhbkz1rxvacxk7m92mb55118391zeworkflowvk97bhbkz1rxvacxk7m92mb55118391zezodvk97axkze7k084r7wta2cnf8rr5800cs1zod-validationvk97axkze7k084r7wta2cnf8rr5800cs1

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments