Implements the three efficiency findings from the #310 pipeline
measurement (refs #35): ~691k subagent tokens / ~3.5 h for a
deliberately small cycle, with recon duplication as the single
biggest obligatory per-cycle cost.
1. Cycle recon (pipeline.md § Cycle recon; specify, planner,
brainstorm, boss, plan-recon): recon fan-out is budgeted per
cycle, not per phase. One full plan-recon dispatch — normally
specify Step 1, in a new pre-spec sources/recon_scope carrier
form — serves both the spec's concrete code shapes and the
plan's file-map. planner Step 2 becomes reuse-first (freshness /
quotability / coverage conditions) with narrowly-bounded delta
dispatches (new Delta dispatch contract in plan-recon.md);
broad Explore-type grounding sweeps are banned in every mode.
Field evidence: 151k (Explore) + 99k (plan-recon) with ~40 %
overlap in one bounded cycle.
2. Fieldtest probe tier (fieldtest, fieldtester, pipeline.md,
agent-template.md § model rule 2): a cycle whose user-visible
delta is a single narrow axis dispatches the per-cycle
fieldtest at tier: probe — 1-2 examples on exactly that axis,
~200-word report, explicit model:sonnet dispatch override (no
dispatch-level effort override exists; frontmatter effort
applies). Probe is a tier, not a skip; the milestone fieldtest
always runs full tier on the frontmatter model, keeping the
milestone-close gate's >=2-example floor unconditional.
3. Assumption economy (specify Step 3 + self-review item 6):
grounding-check cost scales with the spec's assumption count,
so specs state current-behaviour claims only where the change
relies on them (the gate's own falsity test), demote
context-only mentions to ledger citations, and keep iteration
scope tight — fewer restatements, never fewer reliances. The
gate itself is unchanged.
Design reviewed pre-implementation (opus plan review); diff
adversarially verified by a 3-lens review workflow (consistency,
operability, cross-reference), confirmed findings folded in.
refs #35
Effort joins model as a mandatory pin: an omitted field inherits the
session effort, coupling every dispatch's thinking budget to whatever
the user happens to be chatting at (often xhigh) — the same
session-state coupling the model pin removes. The assignment follows
the model split:
- xhigh on every opus agent (judgement roles are the pipeline's
quality floor and must not degrade with the session);
- high on every sonnet agent (tightly-scoped plan execution gains
little from xhigh but pays its latency per dispatch, and these are
the per-task in-loop roles — wall-clock is the efficiency metric;
not lower than high, since re-loops cost more than saved thinking);
- medium inline in the workflow scripts for schema-bound
extraction/verification stages that author no code (preflight,
plan-extract, mini-verify, tree-check, finalize, build/suite
verify).
Workflow agent() calls pass effort explicitly on every call — whether
frontmatter effort propagates through an agentType dispatch is
undocumented, so the scripts do not rely on it. Policy documented in
docs/agent-template.md § effort, mirroring § model.
Agents and workflows previously carried no model field, so every
dispatch inherited the session model — including fable, which is
banned for all plugin agents and workflows by owner decree. Every
dispatch now pins opus or sonnet explicitly.
- opus (low-volume judgment gates whose misses silently poison
downstream work): architect, bencher, debugger, fieldtester,
grounding-check, plan-recon, quality-reviewer, spec-skeptic,
tdd-author
- sonnet (mechanical scope, in-loop or fanned out): docwriter,
glossary-extractor, implementer, spec-reviewer, synthetic-user,
tester
- workflows: all 13 agent() call sites pin a model — sonnet
everywhere except the quality-reviewer gate in implement-loop,
the loop's last correctness check (spec-reviewer only gates
task-text correspondence; real-bug finding is the documented
opus strength)
- docs/agent-template.md: model is now a mandatory frontmatter
field, with the assignment rule and the fable ban recorded
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.
Apply the debug-skill pattern (commits 6410732..de42974) to the
remaining skills: the agent file is the single source for each
interface field's semantics; the SKILL.md copy is reduced to field
names plus a pointer, marking the agent's contract table
authoritative. Because SKILL.md loads into orchestrator context and
agents/*.md into the subagent's fresh system prompt with no
transclusion, duplicated field tables drift apart over time.
- brainstorm: carrier (spec_path/iteration_scope) -> grounding-check;
the absolute-path requirement now lives only in the agent.
- planner: carrier (spec_path/iteration_scope/focus_hint) ->
plan-recon, including the mandatory/optional markers and the
BLOCKED-on-missing rule SKILL.md had omitted.
- fieldtest: carrier + produced fields -> fieldtester; the skill-level
`status` roll-up (clean/friction_found/bugs_found/infra_blocked),
which is not part of the agent's run-status protocol, stays defined
in SKILL.md only.
- docwriter: carrier + produced fields -> docwriter agent.
- implement: carrier (iter_id scratch-dir/stats/not-a-branch
semantics) -> implement-orchestrator; per-task sub-status vocabulary
moved into the orchestrator-agent (it runs the loop in a fresh
context and could not read SKILL.md at runtime, yet referenced "the
sub-status table" by name); task_text_path single-sourced in
implementer with spec-reviewer cross-referencing.
audit was already the reference implementation (pointers, no restated
contracts) and is unchanged.
closes#2
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
plan-recon already had a Non-compile-checked sweep, but it was gated
behind signature/variant/removal scopes (omittable otherwise). The
raw-buf.2 failure was a module addition that shifted a workspace-listing
hash pinned in two test files — no signature change — so the sweep was
legitimately skipped and the twin pin was missed.
Adds "Content-pin-perturbing scopes" to Step 5: when the iteration
shifts a content-addressed/hashed/snapshot value a test pins, the
Non-compile-checked sweep is MANDATORY and must enumerate EVERY test
file that grep-matches the perturbed constant — not just the first.
Updates the omission clause accordingly, leans on the CLAUDE.md
lockstep-invariant pairs (no new profile slot), and adds a Common
Rationalisation plus two Red Flags. Embedded in the agent's own file
so the discipline does not depend on orchestrator memory.
refs #1
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Sixth skill via the boss pattern. planner is the spec→executable-
plan transformation; plan-recon is its read-only file-mapper.
Vocabulary substitutions:
- AILang → "this project"
- milestone → cycle
- Boss → orchestrator
- ailang-plan-recon → plan-recon
- "the architect's Lockstep invariants table" → "the project's
CLAUDE.md lockstep-invariant pairs" (the lockstep pairs are
project-specific; the plugin only carries the pattern that
the orchestrator may declare them)
AILang-specific bits dropped or generalized:
- `docs/specs/<milestone>.md` / `docs/plans/NNNN-<iteration>.md`
→ `paths.spec_dir` / `paths.plan_dir`
- `cargo test --workspace -p <crate>` → "the project's test
command from commands.test, scoped to this test" + generic
template
- `cargo check`, `cargo expand` → "the project's type-checker
/ compiler" and "the project's macro-expansion tool" with
Rust/TypeScript examples in parentheses
- Rust code-snippet template in Step 3 → language-neutral
placeholder with `(exact source code here)` annotations
- Specific past-iter references (loop-recur.1, loop-recur.2,
loop-recur.tidy, remove-mut-var-assign.1, form_a.md,
iter-revert Concern 2, iter effect-doc-honesty Task 2, iter
loop-recur.3 Concern) → generic "recurring defect family
in this project's history" (the discipline survives; the
AILang iter codes that wouldn't parse for a plugin reader
go away)
- `error[E0061]` Rust-specific compiler error code → "hard
compile error" (universal)
Universal substance preserved verbatim:
- Iron Law (3 lines)
- Six-step process (read spec → map structure → write tasks
→ header → self-review → hand off)
- Step 5 self-review checklist — all 8 calibrated checks
(spec coverage, placeholder scan, type consistency, step
granularity, no commit steps, pin/replacement substring
contiguity, compile-gate vs deferred-caller ordering,
verification-command filter strings must resolve)
- All 6 Common Rationalisations rows
- All 6 Red Flags bullets
- plan-recon's nine-step process incl. compile-driven
enumeration (Step 5) and spec-named-path existence table
(Step 6)
- plan-recon's output format including all five sections
(Files, Compile-driven site set, Non-compile-checked
sites, Spec-named-path existence table, Anchors not yet
present)
- plan-recon's 8 Common Rationalisations rows and 9 Red
Flags bullets
Calibrated discipline kept (per memory feedback_dont_strip_
repeated_discipline.md): the three plan-defect families flagged
by user memory (pin/replacement soft-wrap, compile-gate vs
deferred caller, verification filter must resolve) all stay
verbatim — they are project-history scaffolding, just
de-AILang'd in the references.
claude
Brummel