ClawHub

CCW Project Supervisor

v1.0.0

supervise software project execution with openclaude and claude code workflow. use when the user wants an ai supervisor to drive workflow-plan, /issue/plan,...

⭐ 0· 217·0 current·0 all-time
by@majmunu
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 (CCW project supervision) match the included files and runtime instructions: it guides which CCW commands to run, uses provided prompt templates and checklists, and asks for typical project inputs. No unrelated credentials, binaries, or external services are requested.
✓
Instruction Scope
SKILL.md confines runtime behavior to recommending CCW commands, preparing prompts, validating dependency order, and summarizing outputs. It does not instruct the agent to read system config, environment variables, or exfiltrate data to unknown endpoints. It explicitly references only the bundled reference files and CCW commands.
✓
Install Mechanism
No install spec and no code files — instruction-only. Nothing will be downloaded or written to disk by an install step, so installation risk is minimal.
✓
Credentials
The skill declares no environment variables, credentials, or config paths. Its behavior (supervising plans, using templates) does not require secrets or external keys, so the lack of requested secrets is proportionate.
✓
Persistence & Privilege
always is false and model invocation is allowed (platform default). The skill does not request persistent system modifications or access to other skills' configs. There is no elevated or permanent privilege requested.
Assessment
This is an instruction-only supervisor that appears coherent and low-risk. Before installing: ensure you only provide project data you’re comfortable sharing with the agent (avoid pasting secrets, private keys, or CI tokens into prompts), confirm your environment has the CCW/Claude tooling you intend to use, and try the skill in a limited/sandboxed context first. If you want tighter control, consider disabling autonomous invocation or requiring explicit user confirmation before the agent issues any /issue/execute commands in real repositories or task trackers.

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

latestvk978qqtz2hhrws1tecjyq952t982jt61

License

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

Comments