The default project log dir is derived from cwd, so a session run inside a
git worktree resolves to a different slug than the primary clone. Document
that, and point at --project-dir / --file to grade a worktree run.
closes#18
The selector forced every task through the heaviest methodology's
critical path: a behaviour-preserving, type-enumerable change paid the
same specify -> planner -> implement front-half as a novel feature,
because it was neither new behaviour (tdd) nor an observed bug (debug)
and so fell to specify by elimination. Two coupled defects — a selector
with no verification axis, and an all-or-nothing executor — kept the
existing lighter path unreachable and uneconomical. This fixes both.
Part A — verification-keyed selector (boss/SKILL.md):
- Replace the three-way "design line" with an ordered cascade that adds
a verification/enumeration axis ahead of the settled-vs-fork question.
Each lighter arm carries a positive trigger matched by signature, not
reached by elimination.
- New `compiler-driven` arm: a type/signature edit at a definition site
that propagates mechanically. Observe-then-bounce — make the edit,
build, run the suite; clean build AND suite green unchanged commits;
a hole bounces up (specify for a design choice, tdd for discovered
test-specifiable new behaviour); a regression bounces to debug.
- The observed-bug RED-first gate is first in the cascade, so a
mechanical-looking fix cannot bypass it.
- The straddle rule ("add an enum variant") is codified as a rule:
mechanical/forwarding -> compiler-driven; encodes new behaviour ->
tdd/spec; doubt routes up.
- The executor is the elevated inline carve-out plus a shipped workflow,
not a heavy new skill ("the largest concrete win is small").
Part B — Workflow substrate (implement/workflows/):
- implement-loop.js: the per-task loop as a deterministic script. Each
phase (implementer -> spec-compliance -> quality, + tester for E2E) is
a separate top-level agent() call, so a single phase is independently
invokable and inter-phase aggregation/re-loop is code. Retires the
implement-orchestrator agent's inline-role-switch workaround (the four
phase agents survive as the agent-types the script dispatches).
- compiler-driven-edit.js: the observe-then-bounce loop.
- install.sh / uninstall.sh symlink shipped workflows into
~/.claude/workflows/.
- specify and brainstorm stay prose + interactive (human-intent oracle);
only the autonomous/mechanical loops moved. try-and-error is deferred.
Docs (pipeline taxonomy, design, agent-template, migration, README) and
all selector<->executor cross-references updated; the arm and its
executor are co-located so a future re-route through the full loop is a
visible regression.
Verified by an adversarial multi-agent pass: PASS on all six acceptance
criteria; two coherence concerns fixed. The shipped scripts are
syntax-validated but exercised only in a downstream target project (the
skills repo is not itself a pipeline target).
closes#7
The profile was never parsed — it was prose the skill bodies told the model to read, so most slots were dead, constant across every project, or fiction (the whole pipeline block, including the "tdd is opt-in" claim, was enforced by nothing).
Split it in two: constants become fixed conventions named directly by the skills (new docs/conventions.md), and the few genuinely per-project facts move to each project's CLAUDE.md under '## Skills plugin: project facts'. tdd/fieldtest/docwriter are now always available; the only behavioural toggle left is spec auto-sign.
Delete docs/profile-schema.md and templates/project-profile.yml; add docs/conventions.md and a project-facts section to templates/CLAUDE.md.fragment; rewrite all SKILL/agent prose and the pipeline/design/migration/README/INSTALL docs accordingly.
Remove the price-table cost derivation entirely. The transcript carries
no metered cost, and list prices vary by plan — so converting tokens to
a dollar figure dressed an estimate up as a number. Report raw token
counts instead, kept split by class (input / output / cache-creation /
cache-read) with a `total_tokens` sum per scope.
Script: drop DEFAULT_PRICING / load_pricing / rate_for / cost_of and the
--pricing flag; aggregate_transcript and analyze_subagents no longer take
a pricing arg; every `cost_usd` field becomes `total_tokens`; the pricing
warning is gone. SKILL.md: the first graded axis is now "Token spend &
efficiency", the scorecard and output format cite token counts, and the
Iron Law / Red Flags now forbid converting tokens to a cost.
cache_hit_ratio and the active-session warning are unchanged.
The subagent terminal-status classifier scanned the whole final report
text for status keywords, so `\bBLOCKED\b` fired on the implement
end-report's own template lines (`BLOCKED file: BLOCKED.md`,
`Blocked detail:`) and on prose narrating a surmounted blocker. Shipped
implement runs whose end-report reads `Status: DONE` were flipped to
BLOCKED — making the agent-effectiveness axis actively misleading.
Anchor on the structured `Status:` line of the end-report instead (the
implement-orchestrator emits a fixed `Status: DONE|PARTIAL|BLOCKED|...`
header, optionally markdown-bold). Fall back to a status token standing
alone on its own line for agents that emit a bare terminal token; prose
mentions and `BLOCKED.md`-style substrings no longer match. Agents with
no structured status now read `unknown` rather than a fabricated BLOCKED.
Add an executable spec (postmortem/tests/test_terminal_status.py, plain
asserts, no pytest dep) pinning the issue-#3 case plus the genuine
BLOCKED / bare-token / no-signal guards. RED before, GREEN after.
Verified on the real AILang-style session 2811d227: 3 BLOCKED (one a
false positive) -> 2 genuine BLOCKED + 13 honest unknown.
closes#3
A new utility skill that grades a finished session on three axes —
cost & efficiency, toolchain health, agent effectiveness — from the
session's own Claude Code flight recorder (the JSONL transcript plus
one sidechain log per dispatched subagent).
The heavy lifting lives in scripts/postmortem.py, a stdlib-only,
read-only aggregator. It handles the two accounting traps the raw log
sets: usage is repeated per streamed assistant line under a shared
requestId (deduped max-per-field, then summed across requests), and
the transcript carries no costUSD/durationMs (both null) — so cost is
derived from tokens x an overridable list-price table and labelled an
estimate, wall-clock from timestamp deltas. Subagent spend is billed
separately from its own usage objects and graded by terminal status.
Single-session by design: defaults to the newest transcript in the
project's log dir (flagged active if still live), or --session <id>
for a finished run. Secret files are out of bounds in both the Iron
Law and the Red Flags.
Verified end-to-end against a real 15-subagent session.
Brummel