meta-prompting
v2.56.0Enhanced reasoning via /think, /verify, /adversarial, /edge, /compare, /confidence, /budget, /constrain, /json, /flip, /assumptions, /tensions, /analyze, /tr...
⭐ 0· 125·0 current·0 all-time
byIlia Alshanetsky@iliaal
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description promise enhanced reasoning patterns and the SKILL.md contains only pattern definitions and usage. No binaries, env vars, or installs are required, which is proportionate for a purely instructional meta-prompting skill.
Instruction Scope
The instructions tell the agent to auto-trigger patterns when 'context warrants' and to accept new pattern definitions mid-conversation. That gives the skill broad discretion to change agent behavior and output format, which is consistent with a meta-prompting tool but worth noting because it can alter responses across many contexts. The instructions do not request files, credentials, or external endpoints.
Install Mechanism
No install spec or code files — instruction-only. This is the lowest-risk install model and nothing is written to disk or fetched.
Credentials
Requires no environment variables, credentials, or config paths. The SKILL.md does not instruct access to secrets or unrelated system artifacts.
Persistence & Privilege
always:false and user-invocable default; the skill does not request permanent presence or changes to other skills or system-wide settings.
Assessment
This skill is internally consistent and low-risk because it only supplies prompting patterns and asks for no installs or credentials. Before installing, consider: (1) the skill can auto-trigger and accept new pattern definitions mid-session — that can broadly change agent behavior, so avoid enabling it for agents that handle secrets or high-risk actions; (2) test it in non-sensitive contexts to confirm outputs meet your expectations (verbosity, JSON formatting, adversarial turns); (3) if you want tighter control, only invoke it manually and review the SKILL.md to understand which patterns will be applied automatically. If you need the agent to never change behavior autonomously, keep this skill disabled or restrict it to user-initiated use.Like a lobster shell, security has layers — review code before you run it.
latestvk97cjkgjzq4xvmnv6rm4a86pzn84tx7d
License
MIT-0
Free to use, modify, and redistribute. No attribution required.