ClawHub

felt cute uploaded skill

v1.0.0

Orchestrate parallel codebase research. Spawns multiple researcher subagents to investigate different areas, then synthesizes findings into research.md.

⭐ 0· 41·0 current·0 all-time
by@flip-in
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
The name/description (orchestrating parallel codebase research) matches the instructions: decomposing a question, spawning researcher subagents, collecting results, and writing research.md. There are no unexpected binaries, env vars, or config paths requested.
ℹ
Instruction Scope
SKILL.md instructs the agent to read user-specified files (and to use the Read tool without limit/offset) and to spawn Task tool subagents (all calls in a single message). Reading repository files and running git commands is expected for this purpose, but 'use Read tool without limit/offset' could lead to large/unbounded reads of repository content — this is functionally necessary for deep research but is broader than a single-file lookup. Also the document references $ARGUMENTS (platform placeholder) — ensure this maps only to the user-provided question.
✓
Install Mechanism
Instruction-only skill with no install steps, no external downloads, and no code files. This is low risk and consistent with the stated behavior.
✓
Credentials
The skill declares no required environment variables, credentials, or config paths. The only external actions are calls to platform tools (Read, Task) and local git commands, which are proportional to repository research.
✓
Persistence & Privilege
always:false and no special persistence or cross-skill configuration changes requested. Autonomous invocation is allowed (platform default) but not combined with other privilege escalations.
Assessment
This skill appears to do exactly what it says: read repository files, spawn parallel researcher subagents via the platform Task tool, run a few git commands, and write research.md. Before installing, confirm you are comfortable with the agent reading the repository in bulk (the instructions explicitly permit unlimited reads) and ensure $ARGUMENTS maps only to the user-supplied research question. If you want tighter controls, ask for limits on which directories can be read or require explicit confirmation before scanning large paths.

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

latestvk970q8hqzf338348tc754vgf7184vpmj

License

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

Comments