Lintel on GitHub Copilot. Generated from
skills/frontend-typography/SKILL.md; edit the canonical file, then runli-copilot init.
- Resource root:
../../..from this skill's base directory (the Lintel source withbin/,lib/,skills/). Write plans, state and evidence into the working repository's.claude/tree, never into the resource root.- Skill-relative paths: this skill's own
scripts/,references/anddata/folders (and a<base>that the workflow defines as its own directory) mean../../../skills/frontend-typography/in the Lintel source, not this generated folder.${LINTEL_SKILLS_DIR:-skills}means the skills root,../../../skills. Abin/li-runstep runs in the working repository, so use$LINTEL_SKILLS_DIR/frontend-typography/there.- Shell steps: run Bash snippets with Bash (Git for Windows'
bash.exeon Windows, neverSystem32\bash.exe). Save a snippet to a temporary.shfile and runbash "<resource root>/bin/li-run" <file>; it preparesLINTEL_SOURCE_ROOT,LINTEL_REPO_ROOTand the profile context.- Tools: Read=
view, Write=create, Edit=edit, Bash=bash/powershell, Grep=grep, Glob=glob, AskUserQuestion=ask_user, TodoWrite=the plan checklist, Task or a named role=taskwith that custom agent, WebFetch=web_fetch.- Other Lintel workflows are native skills: invoke
/li-<name>rather than reading their files. Named roles such asCodeReviewerare custom agents.
You are the frontend-typography sub-skill — typography-curator for the frontend-design family.
What this skill does
Reads operator brief → TypographyCurator selects font roles from the verified profile,
retained corpus evidence and actual project constraints, then maps variable axes,
size scale, line heights and loading strategy → writes typography.json
(schema_version: 1) with licensing context.
Solo-invokable for component-mode ("just typography please") or auto-invoked by the /li-frontend-design orchestrator in parallel-dispatch (Workflow Step 2).
L-001-discipline: skill body is the contract. Agent at invocation produces specific font choices and licensing-instructions. Don't pre-bake recommendations in the SKILL.md body.
Use the typography definition in the shared design contract. Retain actual profile/brief precedence and source/version/license/rationale evidence for chosen fonts; licensed availability is not established by this example.
When to use
- Solo: "ai-app for legal professionals — give me a typography stack"
- Orchestrator-parallel: dispatched from
/li-frontend-designStep 2 - Brand-update: "customer-deck just landed, what should our heading-stack be?"
When NOT to use
- Already have a typography.json — invoke
/li-frontend-designdirectly with--pattern <name> - Pure palette extraction —
/li-generate-style-learnis right tool (palette ≠ typography) - Font-rendering troubleshooting — that's CSS-debugging, not design-direction
Inputs
- Required
--brief <text>OR--target-audience <description>(one or other minimum) - Optional
--mood <serif-display|tight-mono|variable-experimental|editorial|techy>— override default mood-inference - Optional
--out <path>— output path (default: stdout if solo,$run_dir/typography.jsonif orchestrator-parallel) - Optional
--customer-share— selects the customer-share control boundary; no automatic license validator
Workflow
Step 1 — Parse + warm context
brief="${BRIEF:-${1:-}}"
audience="${TARGET_AUDIENCE:-}"
mood="${MOOD:-auto}"
out="${OUT:-}" # absent --out means stdout, not a path to validate
[ -z "$brief$audience" ] && { echo "Need --brief OR --target-audience"; exit 2; }
Step 2 — Corpus query + TypographyCurator agent dispatch
Query the design corpus first (ADR-0015 — retrieval before generation):
python3 "${LINTEL_SKILLS_DIR:-skills}/design-dna/scripts/search.py" "<mood + audience keywords>" --domain typography -n 3
The verified active design profile's actual font roles are the starting point.
Corpus pairings + the brief justify deviation;
no deviation needed → the profile stack IS the answer. python3 absent → Read
skills/design-dna/data/typography.csv directly (73 pairings, greppable).
Use agents/frontend/TypographyCurator.md as the decision method with the corpus
hits + verified profile. Follow axis ownership;
delegate only when a separate context is useful and actually available. The role
returns a draft; this caller owns Step 3's single publication. Compare candidates
against the actual audience, language/glyph coverage,
heading/body/mono roles, density, available weights/axes, fallback metrics and
project loading constraints. Prefer an already suitable selected font or system
fallback over an unnecessary dependency. A vendor label, installed font or
corpus hit is not a license grant; retain exact source/release evidence.
Use actual supplied/verified license evidence or request an authorized lookup. Neither a role name nor a font recommendation proves current terms were checked.
Step 3 — Produce typography.json
{
"schema_version": 1,
"generated_at": "<iso-8601>",
"brief_summary": "<one-line>",
"mood": "<inferred-or-specified>",
"font_stacks": [
{
"role": "heading",
"family": "<selected heading family>",
"fallback_stack": ["<observed compatible fallback>", "serif"],
"variable_axes": {"weight": [400, 700], "optical_size": [14, 96]},
"loading_strategy": "self-hosted via @font-face",
"license": {
"type": "<verified terms for the selected release>",
"source": "<actual release/file source>",
"operator_instruction": "Verify the selected font release and license, then use an explicitly authorized project asset path and record its source; do not create or scan a personal asset folder"
}
},
{
"role": "body",
"family": "<selected body family>",
"fallback_stack": ["<observed compatible fallback>", "sans-serif"],
"variable_axes": {"weight": [400, 600], "slant": [-10, 0]},
"loading_strategy": "<project-supported, authorized delivery strategy>",
"license": {"type": "<verified terms>", "source": "<actual source>", "operator_instruction": "<required setup and notices; no automatic fetch>"}
},
{
"role": "mono",
"family": "<selected mono family, if needed>",
"fallback_stack": ["monospace"],
"variable_axes": {"weight": [400, 700]},
"loading_strategy": "<project-supported, authorized delivery strategy>",
"license": {"type": "<verified terms>", "source": "<actual source>", "operator_instruction": "<required setup and notices; no automatic fetch>"}
}
],
"size_scale": {
"ratio": 1.25,
"base_px": 16,
"scale": ["xs:0.75rem", "sm:0.875rem", "base:1rem", "lg:1.125rem", "xl:1.25rem", "2xl:1.5rem", "3xl:1.875rem", "4xl:2.25rem", "5xl:3rem", "6xl:3.75rem", "7xl:4.5rem"]
},
"line_heights": {
"tight": 1.1,
"snug": 1.25,
"normal": 1.5,
"relaxed": 1.625,
"loose": 2
},
"letter_spacing": {
"tight": "-0.02em",
"normal": "0",
"wide": "0.05em",
"wider": "0.1em"
},
"operator_instructions_md": "# Typography setup\n\nVerify the actual release/source/license for each selected font. Use an explicitly selected authorized project asset path or approved delivery source; record the reference and required evidence before customer sharing. Do not install, fetch, publish or create personal asset directories from this example."
}
Agent fills in specific choices based on the brief. Don't pre-bake.
Step 4 — Schema-validate + emit
Keep Step 3's actual parsed JSON object as fragment until validation and any
required licensing checks finish. In the trusted source Python scope, repo is
the explicit target root and out is None when --out was omitted, otherwise
the literal repository-relative output path. For named output, the authorized
caller captures original_output_state through P03 before generation (None
means originally absent, not overwrite permission). Then execute:
import sys
from design_contract import emit_fragment
try:
emit_fragment(fragment, "typography", repo=repo, out=out,
original_output_state=None if out is None else original_output_state)
except (ValueError, OSError, UnicodeError) as error:
print(f"ERROR [lintel/design]: {error}", file=sys.stderr)
raise SystemExit(2)
This emits only the validated fragment, not a success-shaped validation receipt.
Invalid data emits no stdout or named file; publication errors have a nonzero exit.
Never pass stdout/special/absolute paths to the rooted-file reader. Its existing
CLI remains valid for an already written, explicitly owned relative file.
For --customer-share, use the customer-share control boundary
on the same data or an owned relative staging file before release. Missing
mandatory licensing/policy evidence remains unverified; stdout is no exemption.
Use its named font-licensing procedure for the exact font releases and intended
use. Required obligations from the brief/profile apply even without this flag.
Reusable patterns
Follow the reusable pattern consumer contract.
Direct entry resolves, or verifies a supplied lock or projected pattern_context, before
choosing. Mandatory clauses bound the typography decision; defaults apply only where the brief
did not decide; unconstrained choices follow the usual Design DNA brief > profile > corpus
rules. Record the clause IDs each choice satisfies; prose clauses need ordinary evidence review.
Pattern text is not evidence of licensing or accessibility.
Status protocol
- DONE — typography.json written, schema valid
- DONE_WITH_CONCERNS — an optional font's terms are unresolved; not permission to use it where licensing is mandatory
- BLOCKED — brief unparsable, OR customer-share license-check failed
- NEEDS_CONTEXT — brief lacks audience-direction
Pause-points
- Customer-share + commercial-license font: surface licensing-instruction explicit + ask for confirm before proceeding
- Brief mentions specific font operator doesn't know about: agent verifies at invocation, surface if unclear
Integration
Reads:
--briefargument- Explicitly selected and authorized licensed-font assets or verified profile references; no personal-home discovery or lazy-created asset folder
Writes:
typography.json(stdout default, $OUT-path if orchestrator)- Audit-log:
.claude/runtime/audit/frontend-typography-runs.jsonl
Calls into:
agents/frontend/TypographyCurator.md(primary)- The
font-licensingsource-inspection procedure in the customer-share control boundary;/li-compliance-gateevaluates its recorded outcomes, not the font license
Consumed by:
/li-frontend-designWorkflow Step 5 (synthesis input)- Operator direct (solo component-mode)
Anti-patterns
- Pre-baking specific font recommendations in SKILL.md body — L-001 violation. Skill body = contract; agent at invocation picks from current font landscape.
- Hardcoding licensing claims — L-003: font licenses change. Agent verifies at invocation.
- Producing typography.json without
schema_version— M-5 compliance.
Failure recovery
- Agent fails to pick (brief too vague): NEEDS_CONTEXT with a specific clarification-question
- Font-recommendation references unavailable font: agent re-picks; logs the attempt
- Schema validation fails: BLOCKED, return diff
Recommended next steps after invocation
- Solo: review typography.json + drop in target project
- Orchestrator: parallel-dispatch returns to
/li-frontend-designStep 5 synthesis - Customer-share: obtain the required
font-licensingevidence, then evaluate the actual controls through/li-compliance-gate; unresolved permission blocks sharing