Credibility Evidence Selector
v1.0.0Pick the strongest credibility evidence for a claim and kill weak evidence chains with the Sinatra Test. Use this skill whenever the user asks 'how do I prov...
⭐ 0· 33·0 current·0 all-time
byHung Quoc To@quochungto
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (selecting and filtering credibility evidence using the Sinatra Test) matches the files and instructions. No unrelated env vars, binaries, or installs are requested and the included reference materials are directly relevant to the stated goal.
Instruction Scope
SKILL.md instructs the agent to ask for or read claim, evidence inventory, and audience context and to rank evidence across six categories and run the Sinatra Test. The instructions only reference prose artifacts and local environment files (claim.md, draft.md, pitch.md, core-message.md) which are appropriate for this task; there is no guidance to exfiltrate secrets, call external endpoints, or run arbitrary shell commands.
Install Mechanism
No install spec and no code files — the skill is instruction-only. This minimizes disk writes and arbitrary-code risk; nothing is downloaded or executed by an installer.
Credentials
The skill requires no environment variables, credentials, or config paths. The only data it needs is the user's claim, evidence inventory, and audience description — which is consistent with its purpose.
Persistence & Privilege
always is false and there are no install-time hooks or requests to modify other skills or system config. The skill can be invoked autonomously by the agent (platform default) but it does not combine that with broad privileges or credential access.
Assessment
This skill is instruction-only and coherent with its purpose: it operates on text you provide (claim, evidence inventory, audience). Before installing/using it, only: (1) avoid supplying sensitive secrets or private credentials as part of the evidence inventory, (2) verify any named customer/authority references yourself before you publish (the skill warns about verifiability and consent), and (3) remember agent outputs are recommendations — review any suggested hero references or rewrites for factual accuracy and permissions before shipping. If you want to prevent autonomous agent invocation of skills generally, toggle the platform setting that controls model-invoked skills (note: that setting is separate from this skill and not required for it to function).Like a lobster shell, security has layers — review code before you run it.
bookforgevk9792b9747g4t802t32grmzec584w7hrcredibilityvk9792b9747g4t802t32grmzec584w7hrevidencevk9792b9747g4t802t32grmzec584w7hrlatestvk9792b9747g4t802t32grmzec584w7hrmarketingvk9792b9747g4t802t32grmzec584w7hrmessagingvk9792b9747g4t802t32grmzec584w7hrpersuasionvk9792b9747g4t802t32grmzec584w7hrproofvk9792b9747g4t802t32grmzec584w7hrsalesvk9792b9747g4t802t32grmzec584w7hrsinatra-testvk9792b9747g4t802t32grmzec584w7hr
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🔬 Clawdis