- Add ai-core/ — Chief, Scout, Scribe OpenClaw agents + shared knowledge base - Consolidate research/reference docs and the sns.md brand foundation under ai-core/knowledge/ - Repoint every sns.md reference (business branding, divisions, root README) to ai-core/knowledge/sns.md - Update root + ai-core READMEs to reflect the new structure
1.5 KiB
1.5 KiB
KNOWLEDGE — Source Standards
How I judge sources and label confidence. This keeps my briefs consistent and honest.
Source tiers (strongest first)
- Primary — the original document, filing, dataset, first-hand account, or official record.
- Official / regulatory — government bodies, courts, standards orgs, entity registries (e.g., Indiana INBiz), regulator filings.
- Reputable secondary — established outlets/experts that cite their sources.
- Tertiary / aggregators — encyclopedias, summaries, "top 10" posts — use for leads, not as proof; follow them back to the primary.
- Anonymous / user-generated — forums, social posts, unattributed blogs — a lead at best; verify independently before use.
Confidence definitions
- Known — directly supported by strong (ideally primary) evidence.
- Probable — best explanation given available evidence.
- Possible — plausible but insufficiently supported.
- Disputed — credible sources genuinely conflict.
- Unknown — evidence doesn't support a responsible conclusion.
Red flags (lower confidence)
- Many articles echoing a single origin (one press release wearing many hats).
- Source has an incentive to mislead.
- Information is stale / no date.
- No primary source is reachable behind the claim.
- Numbers/quotes that no one attributes.
Citing
Link close to the claim it supports; note the source tier and why it's credible; flag date sensitivity when currency matters.