ClawHub

Daily Loop Runner

v1.0.0

Run one controlled daily project loop for a single active project. Use on: scheduled daily runs, planner-initiated project steps, resuming a project after cl...

⭐ 0· 72·0 current·0 all-time
by@sunbinnju-star
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (advance one project one step/day) align with the runtime instructions: state read, bottleneck selection, task construction, execution, and writeback. Nothing requested (no env vars, no binaries, no config paths) appears unnecessary for the stated planning/looping function.
✓
Instruction Scope
SKILL.md confines operations to reading provided inputs (project_card, logs, open_questions) and producing a structured output. It instructs dispatching a task to a selected agent/tool, which is consistent with a planner-style skill; it does not instruct reading system files, environment variables, or contacting external endpoints beyond invoking an agent/tool.
✓
Install Mechanism
Instruction-only skill with no install spec and no code files. Nothing is written to disk and there are no external downloads or package installs, so install risk is minimal.
✓
Credentials
Skill declares no required environment variables, credentials, or config paths. The inputs it requires are logical for its purpose. There are no unexplained secret requests.
✓
Persistence & Privilege
Skill is not always: true and does not request persistent system changes. It permits normal autonomous invocation (platform default) to dispatch tasks to a selected agent; combined with no extra privileges, this is proportionate.
Assessment
This skill is internally consistent and low-risk as provided: it only processes the project_card and logs you feed it and produces a structured plan step. Before installing, confirm that the project_card and daily logs you supply contain no sensitive credentials or secrets (the skill will read full project state). Also review which agents/tools the skill may dispatch to — those agents' permissions determine what actions can actually be taken. If you intend to allow autonomous action, limit the connected agents' scopes/permissions and monitor writeback logs to ensure expected behavior.

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

latestvk971qerh7k59yngn6mddnk04ch83qhwm

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments