bassclef
About

Built with bassclef

A running tally of the disciplines that shipped this site — luminaries dispatched, skills fired, temperance-per-PR, and the negative-space signals that separate honest data from vanity metrics.

Every number below is a count computed from files in this repository. Skills and luminaries leave marker files as they run; the tally is regenerated on every push to main.

Headline metrics 30-day window

Why this matters: Sam wants trust proof he can absorb in 30 seconds — the top row answers “is the discipline actually running?”. Louis wants a fair read of adoption cost — the ratios say what's cheap vs. what's not. Morgan wants raw counts to check the abstraction is earned — every number is a file path away.

Temperance per PR

97.7%

42 of 43 PRs merged this month had a temperance marker on their branch.

Chronicles written

22

One narrative log per substantive session; 39 lifetime.

Luminaries dispatched

10 distinct

alan-cooper, michael-feathers, sophia-prater top the frequency.

Skills dispatched

50 distinct

/docs most frequent.

Markers written

134

Across 18 gate types; audit trail per commit.

Data freshness

partial

Generated 8/30/2026, 9:12:47 PM.

Cost signal illustrative · token-only

Why this matters: Sam wants a rough sense of what running bassclef costs. Louis wants to project his own bill before adopting. Morgan wants the underlying math not the marketing. Token estimate is illustrative — actual per-turn tokens vary by model + task shape; hosting + CI + tools not included.

Approximate turns

1,380

Sum of chronicle turns field across 13 chronicles this window. 9 more chronicles missing the field — real count is higher.

Estimated tokens

~2.07M

approximate_turns × 1,500 tokens/turn (rough average for /longrun sessions). Illustrative — actual varies by model, tool use, and prompt shape.

Models used

Claude Opus 4.7 (per session config)

From chronicle frontmatter model field where present, else session-config default.

Luminaries dispatched 30d · top 10

Why this matters: Louis picks lenses relevant to his org's work — Cooper for marketing surfaces, Beck for infrastructure, Norman for interaction. Which luminaries lead here signals what kind of work shipped. Sam sees named practitioners, not marketing terms.

SlugDispatchesTrend
alan-cooper2
michael-feathers2
sophia-prater2
edward-tufte2
don-norman1
david-parnas1
douglas-atkin1
donald-norman1
frederick-brooks1
tony-hoare1

Skills dispatched 30d · top 15

Why this matters: Morgan checks which gates actually fire vs. which live only in the README — dispatch counts show what earns its keep. Sam sees which named workflows he'd meet on day one.

SkillDispatches
/docs39
/longrun19
/build11
/ticket8
/launch6
/lean-canvas6
/session-end6
/howdoi6
/auth6
/download6
/gone6
/diagnose5
/roadmap-reconcile5
/skill5
/riff5

Negative space honest posture per Popper

Why this matters: Morgan and Louis both flinch at frameworks that only report their wins. Popper's falsifiability principle says the claim is only meaningful if the counter-signal is visible too. Sam reads negative space as the strongest single trust signal on the page.

What the framework did NOT do — the counter-signal that separates honest data from vanity metrics.

20 of the last 22 chronicles missing the authoring_luminaries frontmatter field (91%). Discipline holds unevenly; the framework surfaces its own gaps.

1 PRs merged this month WITHOUT a temperance marker. Discipline is not perfect; honest signal beats vanity metric.

Known open issues on this site

  • bassclef-web#202 — agents emit retired WU-N vocab in new-authored plan docs and prep output (fix in flight)
  • bassclef-web#203 — longrun-prep compounding-axis hook fires false-positive on compliant 3-option prep (fix in flight)
  • bassclef-web is private during pre-release; opens publicly at v1.0. Issue numbers referenced by ticket ID.