Fact Check
Verify technical claims against primary sources — docs, source code, changelogs, reproducible examples — and mark what couldn't be verified instead of masking it with a confident tone. Use when you say "fact-check this", "verify these claims", doubt a technical statement, or before shipping material with disputed claims.
Install
$ agentstack add skill-archiesa-skills-fact-check ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-archiesa-skills-fact-check) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Fact-check against primary sources
Procedure
- Pull every statement that reads as a technical fact: API behavior, numbers, comparisons, versions, limits.
- Classify each: fact / personal experience / hypothesis. Experience and hypotheses must be marked as such ("in my experience", "seems like") — they can't read as law.
- For each fact, find a primary source: official docs, source code, changelog, RFC, issue, or a reproducible example. A secondary retelling (a blog, a chat) is a lead to the source, not proof.
- Verdict per claim: confirmed (with the source) / refuted (and what's actually true) / couldn't verify.
- Fix: correct what's refuted; soften the unverified to experience or hypothesis, or cut it; reframe an unproven "X is faster than Y" into the logic of a concrete scenario.
- Report back: claim → verdict → source. Name the unverified plainly — don't mask it with a confident tone.
What to check with
- Web: official docs, the project repository, changelogs, RFCs.
- Code: a short reproducible example right here (a REPL, a sandbox) — often faster and more reliable than searching.
Red flags
- "always / everywhere / in all versions" — often the residue of compression that dropped the reason.
- A pretty claim you'd rather not check — check it first (confirmation bias).
- A number with no measurement method behind it.
- A comparison ("X is cheaper / faster than Y") stated as a general law from a single case.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ArchieSA
- Source: ArchieSA/skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.