M126 — ${CLAUDE_EFFORT} + /skills filter

First M126 bundle: #1540 (CC 2.1.120 env var) and #1546 (CC 2.1.121 type-to-filter). Pure SKILL.md doc updates — no new code, no MCP risk.
CC 2.1.120 CC 2.1.121 files 4 prod code unchanged

#1540 — Pick an effort, see what each skill does

${CLAUDE_EFFORT} (set by `/effort` or model picker)
/ork:assess
Quality scoring 0-10 across N dimensions
dimensions:
uncertainty pass:
time:
/ork:explore
Multi-angle codebase exploration
parallel agents:
phases:
time:
/ork:cover
Test suite generation + heal loop
tiers:
heal iterations:
time:

How the skills detect effort (CC 2.1.120+)

# Read env first (CC 2.1.120+), then check explicit override
EFFORT = os.environ.get("CLAUDE_EFFORT")  # "low" | "medium" | "high" | "xhigh" | None
for token in "$ARGUMENTS".split():
    if token.startswith("--effort="):
        EFFORT = token.split("=", 1)[1]   # explicit override wins
EFFORT = EFFORT or "high"  # default when CC < 2.1.120 and no flag

Backwards-compatible: on CC < 2.1.120 the env var is unset, so the explicit --effort=... token is the only path. /ork:doctor warns when xhigh is requested without Opus 4.7.

#1546 — Type-to-filter the native /skills picker (CC 2.1.121)

Mock — type below to see how the new native picker filters as you type. /ork:help already pointed users at this in its leading tip.

Decision split

You want to…UseWhy
find one skill quickly/skillstype-to-filter is fastest for known names
browse by category/ork:helpshows BUILD / GIT / PLAN / MEMORY / QUALITY etc.
understand why a skill applies/ork:helpshows trigger keywords + examples