Start Here — Colour Memory API

1

Authentication

Your demo key

No signup. Rate-limited to 50 calls — enough to try everything below. Studio (£12/month) removes the limit and adds full archive access.

API Key

cm_demo_colourmemory_website_internal

Pass as header: X-Api-Key: cm_demo_colourmemory_website_internal

2

Tool 1 of 3

query_hex

Match any hex to the archive. Returns the nearest named colour with its primary source, confidence score, claim_role, and do_not_say guardrails.

curl

curl -X POST https://api.colourmemory.com/query/hex \
  -H "X-Api-Key: cm_demo_colourmemory_website_internal" \
  -H "Content-Type: application/json" \
  -d '{"hex": "#D4A829", "n_results": 1}'

Response — key fields

nameRa Gold Divine Flesh

archiveAncient Egypt

primary_sourceBook of the Celestial Cow; Tutankhamun death mask, Cairo JE 60672

confidence0.88

claim_strengthB — Named documentary source

claim_rolesupport

do_not_say

Do not describe this hex as spectrophotometrically measured — values are computed from material descriptions.

claim_role tells you how hard you can push this entry: anchor = load-bearing evidence. reject = stub, exclude from deliverables.

3

Tool 2 of 3

archive_search

Full-text keyword search across archive colours. Search for "arterial blood" not "red" — this is a text search, not a colour search. Use query_hex for hue-based retrieval.

curl

curl -X POST https://api.colourmemory.com/archive/search \
  -H "X-Api-Key: cm_demo_colourmemory_website_internal" \
  -H "Content-Type: application/json" \
  -d '{"query": "arterial blood", "archive": "Keats", "include_full": true}'

Response

nameKeats's Lung

hex#C8B6B0

primary_sourceThe Letters, 3 February 1820

claim_roleanchor

Always save the slug from results. Pass it to colour_card — slugs never break on apostrophes.

4

Tool 3 of 3 · Commercial palette analysis

brand_audit

Pass a palette of hex values. Returns WCAG accessibility matrix, cultural risk notes by market, archive matches with provenance, palette role assignments, and a commercial verdict. Fully deterministic — no LLM.

curl

curl -X POST https://api.colourmemory.com/brand/audit \
  -H "X-Api-Key: cm_demo_colourmemory_website_internal" \
  -H "Content-Type: application/json" \
  -d '{
    "palette": ["#D4A829", "#1A5C6E", "#24214E", "#0A0A0B"],
    "markets": ["UK", "China", "Middle East"],
    "context": "luxury packaging"
  }'

Response — key fields

wcag_matrixAll foreground/background combinations with AA/AAA pass/fail

cultural_riskPer-market flags with context notes

commercial_labelArchive-verified name for each hex

verdict_copyPlain-English palette verdict

5

No code required

Connect to Claude or ChatGPT

All tools load instantly via MCP. One URL. No SDK, no boilerplate.

https://api.colourmemory.com/mcp

Claude (web)

Settings → Integrations → Add Integration → paste URL

Claude Desktop

Add to claude_desktop_config.json:

"colour-memory": {
  "type": "sse",
  "url": "https://api.colourmemory.com/mcp",
  "headers": {
    "X-Api-Key": "YOUR_KEY"
  }
}

ChatGPT / any client

Settings → Connectors → Add → paste URL. Tools use underscore notation: archive_search, brand_audit, brand_collision

6

Reference

All tools

Start with the three above. Come back to the rest when you need them.

query_hex ★

Match any hex to the archive with CIEDE2000 distance

archive_search ★

Full-text search across archive colour names and notes

brand_audit ★

WCAG matrix, cultural risk, palette roles, verdict

brand_collision

Can this brand own this colour?

archive_evidence_gap

Is this claim supported?

archive_coverage_gap

Which themes are thin?

query_conceptual

Semantic search by mood or concept

ecommerce_namer

Archive names for up to 40 SKUs

colour_card

Full provenance by slug

palette_concept

Heritage palette from cultural concept

archive_report_brief

Editorial research package

index_resonance

Material-consequence score

Full documentation Pricing →

!

Watch out

Common mistakes

✗ Searching for colour by hue with archive_search. Use query_hex or query_conceptual instead. archive_search is text matching, not perceptual matching.

✗ Ignoring claim_role. Anything marked reject is a stub with no primary source. Don't use it in client-facing work.

✗ Using names instead of slugs. Names break on apostrophes. Slugs always work.

✗ Ignoring do_not_say fields. They stop your agent making claims the archive doesn't support. Read them.

✗ Assuming strict_sources filters by evidence grade. It only checks a source is present, not that it meets a quality bar. Pass minimum_claim_strength (e.g. "B") on archive_report_brief or session_brief for a hard, honestly-enforced floor.

✗ Trusting a coverage grade at face value. Check semantic_match on archive_coverage_gap's best_match — it flags when the top lexical result doesn't actually address your theme, even on a "moderate" or "strong" grade.

✗ Multi-word avoid terms. "arsenic pigments" correctly excludes any entry containing both words anywhere in its text, not just that exact phrase — so it will catch "Arsenic Teal Wallpaper" as intended.

?

Transparency

What runs without an LLM?

Most core endpoints are fully deterministic. Some optionally call an LLM for written output — those are labelled clearly in the docs.

Deterministic (no LLM)

Nearest-colour matching (CIEDE2000)

Lab / LCh metrics and LRV

WCAG contrast and accessibility matrix

Palette role assignment

Cultural risk scoring

Image palette extraction

Design token export

Uses LLM (opt-in)

Brand reports and editorial briefs

session_brief forensic narrative

colour_hooks and colour_story

agent_brief for image generation

Colour intelligence
with receipts.

Your demo key

query_hex

archive_search

brand_audit

Connect to Claude or ChatGPT

All tools

Common mistakes

What runs without an LLM?

The archive is ready when you are.

Colour intelligencewith receipts.

Your demo key

query_hex

archive_search

brand_audit

Connect to Claude or ChatGPT

All tools

Common mistakes

What runs without an LLM?

The archive is ready when you are.

Colour intelligence
with receipts.