docs(voyage): document v4.1 profiles + observability + doc-consistency-pinning

Step 22 of v4.1 — write top-level docs for the v4.1 feature surface.

Files updated:
  CLAUDE.md       — Commands tables: add --profile to all 6 modes
                    + new ## Profile system + ## Observability sections
  README.md       — per-command Modes tables: add --profile row
                    + new top-level ## Profile system + ## Observability
                    + cross-link from ## Cost profile
  CHANGELOG.md    — new "## v4.1.0 — 2026-05-09" entry per Keep-a-Changelog 1.1.0
                    (Added / Changed / Fixed / Notes)
  docs/profiles.md — NEW: 168-line decision tree, lookup precedence,
                    custom-profile authoring, drift detection,
                    cost-estimation table with disclaimer
  tests/lib/doc-consistency.test.mjs — extend with 5 new pinning tests:
                    CLAUDE.md --profile + phase_models canonical name,
                    README.md --profile coverage (≥ 6 mentions),
                    CHANGELOG.md v4.1.0 entry, docs/profiles.md substantive

ROADMAP.md is gitignored per marketplace policy (sesjonsfiler) — local
edit applied for v4.1 DONE marker, not committed.

Plan-critic Blocker 2 split is honored: Step 21 pinned commands-only;
Step 22 writes the docs and pins them. doc-consistency.test.mjs is
green AFTER Step 22 (would have failed if Step 22 ran in same wave).

Tests: 487 pass + 2 skipped (Docker not installed).
This commit is contained in:
Kjell Tore Guttormsen 2026-05-09 10:09:44 +02:00
commit f2f8246e01
5 changed files with 394 additions and 0 deletions

View file

@ -0,0 +1,169 @@
# Profile system — voyage v4.1
This document describes the model profile system: built-in tiers,
lookup precedence, custom-profile authoring, drift detection, and
cost estimation (with disclaimer).
## Built-in profiles
Three pre-defined tiers ship with v4.1, located at
`lib/profiles/{economy,balanced,premium}.yaml`.
| Profile | Brief | Research | Plan | Execute | Review | Continue | Use case |
|---------|-------|----------|------|---------|--------|----------|----------|
| `economy` | sonnet | sonnet | sonnet | sonnet | sonnet | sonnet | Lowest cost; small-scope tasks where you have high confidence the brief is right |
| `balanced` (default) | sonnet | sonnet | opus | sonnet | opus | sonnet | Default — opus where reasoning depth pays off (plan synthesis + adversarial review) |
| `premium` | opus | sonnet | opus | sonnet | opus | sonnet | Critical-path planning + review when budget allows |
`balanced` is the v4.1 default. It puts opus on the two phases where
quality matters most (Plan synthesis + Review) and sonnet everywhere
else. This lands the cost/quality trade-off that solo-developers and
small teams actually want.
`economy` is *strictly experimental* in v4.1. The cross-tier Jaccard
floor (0.55) is grounded in parked-synthetic fixtures, not empirical
runs (Step 17 calibration was deferred — see
`tests/synthetic/profile-jaccard-calibration.md`). If you observe
economy-plan quality regressions, fall back to `balanced`.
## Decision tree
```
Are you uncertain whether the brief is correctly framed?
└── Yes → premium (opus on brief + plan + review)
└── No → continue
Is the change small (≤ 5 steps in the plan)?
└── Yes → economy (sonnet everywhere)
└── No → balanced (opus on plan + review)
Special cases:
- Critical-infrastructure plan → premium
- Migration with rollback risk → premium
- Research-heavy task (≥ 4 dimensions) → balanced (research-stage benefits)
- Bug fix with clear reproducer → economy
- Documentation-only PR → economy
```
## Lookup order
Voyage resolves the profile in this priority order:
1. **Explicit `--profile <name>` flag** — passed to the command
2. **Plan-file frontmatter `profile:`** — when resuming via
`/trekexecute --resume` or `/trekcontinue`
3. **`VOYAGE_PROFILE` environment variable** — useful for headless CI
4. **Default `balanced`** — final fallback
The resolved value is recorded in two places:
- Plan-file frontmatter `profile: <name>` and `phase_models: [...]`
- Stats stream `${CLAUDE_PLUGIN_DATA}/trek*-stats.jsonl`
`profile`, `profile_source`, `phase_models`, `model_used`,
`phase_models_resolved` fields
`profile_source` distinguishes how the profile was resolved (`flag` /
`plan_frontmatter` / `env` / `default`), so dashboards can surface
unexpected env-var inheritance in CI.
## Custom profiles
Drop a YAML file at `lib/profiles/<name>.yaml` to define a new tier.
The validator (`lib/validators/profile-validator.mjs`) enforces:
- Every `phase_models[].phase` must be a known phase enum:
`brief` / `research` / `plan` / `execute` / `review` / `continue`
- Every `phase_models[].model` must match `^(opus|sonnet)(\b|-).*` or
one of the canonical short names
- All six phases must be present (no partial profiles)
Custom profiles override built-ins of the same name (lookup is
alphabetical with `<custom>` taking precedence). You may NOT redefine
`balanced` (the default tier is locked to prevent accidental override
of headless CI behaviour); use a different name and reference it via
`--profile <new-name>` or `VOYAGE_PROFILE=<new-name>`.
### Example custom profile
```yaml
# lib/profiles/critical.yaml — opus everywhere except continue
phase_models:
- phase: brief
model: opus
- phase: research
model: opus
- phase: plan
model: opus
- phase: execute
model: opus
- phase: review
model: opus
- phase: continue
model: sonnet
```
Validate with: `node lib/validators/profile-validator.mjs --json lib/profiles/critical.yaml`
## Drift detection
In `--strict` mode, `plan-validator.mjs` emits a `MANIFEST_PROFILE_DRIFT`
warning when the plan-level `profile:` differs from any step manifest's
`profile_used`. The warning is a *signal*, not a failure — the plan
remains `valid: true`. This catches:
- Manual edits where an operator changed a single step's profile
- Resume from a partial run where the previous session used a different
tier
- Copy-paste errors when stitching plan fragments
To suppress the warning intentionally (e.g. when a critical step
genuinely needs a higher tier), document the override in the step's
prose and re-run with `--soft` to validate without strict-mode warnings.
## Cost estimation
> **Disclaimer:** the table below is an *anslag*, not a contractual
> SLA. Real cost depends on context size, agent-swarm cardinality,
> tool-use density, and Claude Code billing schedule. Treat these as
> rough order-of-magnitude.
| Profile | Brief | Research | Plan | Execute | Review | Total |
|---------|-------|----------|------|---------|--------|-------|
| `economy` | $0.100.50 | $0.502.00 | $0.502.00 | $1.005.00 | $0.201.00 | **$210** |
| `balanced` | $0.100.50 | $0.502.00 | $1.004.00 | $1.005.00 | $0.502.00 | **$314** |
| `premium` | $0.502.00 | $0.502.00 | $1.004.00 | $1.005.00 | $0.502.00 | **$415** |
Numbers are per *full pipeline run* (brief + research + plan +
execute + review) on a moderate-complexity task. Numbers scale roughly
linearly with the size of the resulting plan (10 steps ≈ baseline; 30
steps ≈ 3× the execute column).
Per-profile actuals are emitted to JSONL stats — pipe them through the
OTel export (`docs/observability.md`) to get real cost-attribution
graphs in Grafana. Replace the table above with your own measured
numbers after ≥ 3 runs of each profile.
## Deferred to v4.2
- **`balanced.external_research_enabled` operator-override** —
v4.1 omits this per scope-guardian SG2. v4.2 may add an opt-in
flag to enable external research agents in the balanced tier
without forcing premium.
- **Empirical Jaccard re-calibration** — parked-synthetic fixtures
in v4.1 use a 0.55 conservative starting threshold. v4.2 plans an
empirical re-run with $60-120 LLM budget to derive a calibrated
threshold from real economy-vs-premium plan pairs.
- **ROUGE-L + char-4gram MinHash** as primary/secondary cross-tier
gates per research/02 Recommendation #7. Jaccard remains the gate
in v4.1; v4.2 may layer ROUGE-L on top.
## See also
- [`README.md` § Profile system](../README.md) — top-level overview
- [`CLAUDE.md` § Profile system](../CLAUDE.md) — internal reference
- [`docs/observability.md`](observability.md) — JSONL → OTel pipeline
- [`tests/synthetic/profile-jaccard-calibration.md`](../tests/synthetic/profile-jaccard-calibration.md)
— calibration status and threshold rationale
- [`lib/profiles/`](../lib/profiles/) — built-in profile YAMLs
- [`lib/validators/profile-validator.mjs`](../lib/validators/profile-validator.mjs)
— schema validator with CLI shim