Positioning: sharpen vs-context7 security-by-construction + free/unmetered advantages (#52)#53
Merged
Conversation
…ee/unmetered (#52) positioning.md already carried these axes but understated + un-cited, and line 52 over-asserted "queries hit their servers" — context7's Enterprise page contests it (local processing + SOC-2 + self-host). Sharpened + honesty-corrected: - Differentiator #2 → "Private and free, by construction" with inline concession. - Head-to-head: honest Architecture ("hosted SaaS by default") + Privacy ("default routes through their cloud") rows; Access → Cost (unmetered vs metered + paid). - Prose: context7-cited evidence + concessions + dated pricing capture (2026-07-05) + "cite the structure, not the rotting figure" caveat. Both axes are STRUCTURAL, not harness-measured like freshness — so the honesty mechanism is dated citation + concession. Scope: positioning.md only (zero #50 overlap).
…", not "never leaves your machine" (#52) Self-caught honesty fix (#27/#29): LiveDocs DOES fetch from npm/PyPI/docs hosts, so the library name in a query reaches the origin host. The literal "queries never leave your machine" over-claimed. Precise + defensible: no third-party INDEX/vendor sees the query (context7's exact role), and version reconciliation reads the local install offline.
R1 opus ensemble: delivery confirmed, but the framing counted what LiveDocs
REMOVES (vendor / index / meter) without what it ADDS. Closed:
- "nothing to trust or breach" over-claimed → trust doesn't vanish, it SHIFTS to
the local signed binary + its GitHub-Release / --self-update distribution (a
compromised release / signing key is arguably a larger blast radius).
- Reconcile context7's own "queries processed locally" claim → even client-side
processing sends the index lookup to their hosted per-org DB; that hosted index
is exactly what LiveDocs structurally lacks (pre-empts the strongest rebuttal).
- "no third-party index sees your query" narrowed → the origin host still sees it;
the win is removing one observer (context7's index), not "nobody sees it".
- Cost: "unmetered" isn't "costless" (you bear origin-host rate limits + bandwidth);
"private repos" is a different mechanism (local installed copy vs cloud parse),
not a free superset.
- Symmetric conclusion: a DIFFERENT trust surface, not a strictly smaller one.
- Accuracy: drop unverified "/mo" on Free ("1,000 included calls"); SOC-2 Type II
uniformly; source's word "private" DB; self-host = the (custom-priced) Enterprise tier.
Scope: positioning.md only (zero #50 overlap).
…R2 (#52) R2 found a HIGH consistency straggler: #2/table/bullets were sharpened to "origin host still sees it / different trust surface", but the WEDGE (the lead) still asserted flat "private" — a direct contradiction. Propagated the verified register to the lead: - wedge "it is private" → "no third-party doc index in the query path"; lead outcome "(latest, private, keyless)" → "(latest, keyless, no index in the middle)". - "context7 can't become local, private, keyless without giving up its model" → states the reconciliation: centralized SaaS by default; local+keyless = its Enterprise self-host tier (aligns with the self-host concession, drops the absolutist "can't"). - factual hedge: "client-side processing still sends the index lookup" → "a hosted SaaS with a per-org database must still route the index lookup" (architectural inference, not a claim of context7 internals). The anti-positioning "lost privacy" (line ~119) is left as-is — it refers to the third-party index a central cache would add, the one privacy LiveDocs retains.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Refs #52
Summary
Sharpen two vs-context7 advantages in
docs/positioning.md(the external-messaging source of truth) with context7's own marketing as evidence, and honesty-correct an over-asserted line.Changes (
docs/positioning.mdonly)Honesty (#27/#29)
These axes are structural + citable, not harness-measured (unlike freshness) — so the guard is dated citation + concession, not a drift-guarded number. No claim that context7 is "insecure" or "paywalled"; the differentiators are no-vendor-by-construction and unmetered.
Scope / conflict
docs/positioning.mdonly — zero overlap with in-flight #50 (which owns the README sentinel block). Both merge in any order. README blurb + wiki Home mirror deferred as follow-ups.Checklist
Refs #52)/idd-close🤖 /idd-all · no close trailer — manual /idd-close after merge.