Skip to content

code-yeongyu/oh-my-openagent / packages/shared-skills/skills/ulw-research

ulw-research

Unverified

Runs maximum-saturation research with a cooperating team, claim-graph gating, and a cited, QA'd deliverable. Use when the user explicitly asks for research or a deep investigation, including any 'ulw' research wording.

FileKindSize
AGENTS.mdmarkdown9.2 KB
ATTRIBUTION.mdmarkdown2.4 KB
SKILL.mdskill md42.2 KB
references/deliverable-phase.mdmarkdown49.1 KB
references/fact-ledger-grammar.mdmarkdown6.8 KB
references/fact-ledger.mdmarkdown24.4 KB
references/fidelity-claim-kinds.mdmarkdown5.2 KB
references/markdown-blocks.mdmarkdown721 B
references/report-gates.mdmarkdown32.0 KB
references/report-visualization.mdmarkdown15.4 KB
references/research-collection.mdmarkdown26.8 KB
references/research-delivery.mdmarkdown14.3 KB
references/unit-rules.mdmarkdown16.3 KB
scripts/claim-kinds.mjsscript11.2 KB
scripts/claim-kinds.test.tsscript13.5 KB
scripts/claim-prefix.test.tsscript4.9 KB
scripts/claims-index.mjsscript5.3 KB
scripts/claims-index.test.tsscript5.7 KB
scripts/cli-support.mjsscript4.1 KB
scripts/cli-support.test.tsscript4.3 KB
scripts/contracts-validation.test.tsscript3.3 KB
scripts/contracts.mjsscript26.1 KB
scripts/contracts.test.tsscript7.9 KB
scripts/conversion-altered.test.tsscript11.0 KB
scripts/conversion-originals.test.tsscript22.7 KB
scripts/conversion-probe-classes.test.tsscript45.2 KB
scripts/conversion-probe.mjsscript13.5 KB
scripts/conversion-probe.test.tsscript11.6 KB
scripts/copy-returns-fidelity-policy.test.tsscript10.4 KB
scripts/copy-returns-polish.test.tsscript12.1 KB
scripts/copy-returns.test.tsscript10.6 KB
scripts/css-hiding.mjsscript9.1 KB
scripts/css-lite.mjsscript7.6 KB
scripts/css-lite.test.tsscript6.4 KB
scripts/delivery-barrier-meaning.test.tsscript12.3 KB
scripts/delivery-barrier.test.tsscript11.4 KB
scripts/design-spec.mjsscript7.2 KB
scripts/design-spec.test.tsscript6.7 KB
scripts/entities.mjsscript2.5 KB
scripts/entry-guard.mjsscript794 B
scripts/fact-commands-fill.mjsscript6.6 KB
scripts/fact-commands-meaning.mjsscript12.6 KB
scripts/fact-commands-polish.mjsscript10.3 KB
scripts/fact-commands-verify.mjsscript12.6 KB
scripts/fact-commands.mjsscript13.8 KB
scripts/fact-derive-2.test.tsscript7.2 KB
scripts/fact-derive.mjsscript4.9 KB
scripts/fact-derive.test.tsscript10.8 KB
scripts/fact-expr.mjsscript4.5 KB
scripts/fact-fill.mjsscript11.7 KB
scripts/fact-fill.test.tsscript10.7 KB
scripts/fact-ledger-grammar.test.tsscript11.8 KB
scripts/fact-ledger-kinds.mjsscript16.5 KB
scripts/fact-ledger-kinds.test.tsscript9.3 KB
scripts/fact-ledger-legacy.test.tsscript8.9 KB
scripts/fact-ledger-schema.mjsscript11.9 KB
scripts/fact-ledger-schema.test.tsscript1.5 KB
scripts/fact-ledger.mjsscript5.5 KB
scripts/fact-ledger.test.tsscript9.7 KB
scripts/fact-locate.mjsscript12.1 KB
scripts/fact-locate.test.tsscript7.2 KB
scripts/fact-pairs.mjsscript8.3 KB
scripts/fact-pairs.test.tsscript6.6 KB
scripts/fact-recompute.mjsscript4.8 KB
scripts/fact-unit-dimensions-2.test.tsscript8.4 KB
scripts/fact-unit-dimensions.test.tsscript11.9 KB
scripts/fact-unit-rules.test.tsscript8.9 KB
scripts/fidelity-binding.mjsscript6.7 KB
scripts/fidelity-binding.test.tsscript4.0 KB
scripts/fidelity-brief.mjsscript6.4 KB
scripts/fidelity-brief.test.tsscript7.8 KB
scripts/fidelity-claims.mjsscript5.1 KB
scripts/fidelity-csv.mjsscript1.1 KB
scripts/fidelity-lexicon.mjsscript8.1 KB
scripts/fidelity-lexicon.test.tsscript6.0 KB
scripts/fidelity-originals.mjsscript9.9 KB
scripts/fidelity-policy.mjsscript10.1 KB
scripts/fidelity-records.mjsscript13.4 KB
scripts/fidelity-snapshot-readability.mjsscript2.2 KB
scripts/fidelity-sources.mjsscript4.3 KB
scripts/fidelity-state.mjsscript2.1 KB
scripts/fidelity-state.test.tsscript1.7 KB
scripts/fidelity-trace.mjsscript6.1 KB
scripts/fidelity-trace.test.tsscript3.3 KB
scripts/fidelity-value-unit.mjsscript13.3 KB
scripts/fidelity-values.mjsscript16.1 KB
scripts/fidelity-values.test.tsscript15.8 KB
scripts/format-extract-css.mjsscript5.1 KB
scripts/format-extract.mjsscript10.3 KB
scripts/format-extract.test.tsscript11.2 KB
scripts/gates-context.test.tsscript2.6 KB
scripts/gates-fidelity.mjsscript6.3 KB
scripts/gates-fidelity.test.tsscript3.3 KB
scripts/gates-figures.mjsscript9.9 KB
scripts/gates-layout.mjsscript6.6 KB
scripts/gates-layout.test.tsscript12.4 KB
scripts/gates-static-headings.test.tsscript961 B
scripts/gates-static.mjsscript8.1 KB
scripts/gates-static.test.tsscript14.9 KB
scripts/gates-structure.mjsscript4.7 KB
scripts/gates-text.mjsscript7.2 KB
scripts/glossary-vocabulary.test.tsscript1.7 KB
scripts/html-foreign-object.mjsscript2.8 KB
scripts/html-lite.mjsscript9.7 KB
scripts/html-lite.test.tsscript7.9 KB
scripts/html-markup.mjsscript7.0 KB
scripts/html-text-map.mjsscript4.8 KB
scripts/html-text-map.test.tsscript15.2 KB
scripts/layout-probe.mjsscript4.6 KB
scripts/layout-probe.test.tsscript4.2 KB
scripts/lexicon.mjsscript2.6 KB
scripts/lexicon.test.tsscript3.2 KB
scripts/markdown-closers.mjsscript4.9 KB
scripts/md-lite-boundaries.test.tsscript2.2 KB
scripts/md-lite.mjsscript7.8 KB
scripts/md-lite.test.tsscript10.3 KB
scripts/meaning-check.mjsscript9.7 KB
scripts/meaning-check.test.tsscript12.0 KB
scripts/number-grammar.mjsscript2.1 KB
scripts/number-grammar.test.tsscript1.8 KB
scripts/numerals.mjsscript12.0 KB
scripts/numerals.test.tsscript6.1 KB
scripts/outcome-start.mjsscript3.8 KB
scripts/outcome-start.test.tsscript5.0 KB
scripts/outcome.mjsscript10.5 KB
scripts/outcome.test.tsscript12.8 KB
scripts/page-source-locator.test.tsscript11.7 KB
scripts/particles-ko.mjsscript4.8 KB
scripts/particles-ko.test.tsscript4.3 KB
scripts/placeholder.mjsscript9.3 KB
scripts/placeholder.test.tsscript3.8 KB
scripts/polish-apply.mjsscript6.8 KB
scripts/polish-apply.test.tsscript19.7 KB
scripts/polish-diff.mjsscript1.7 KB
scripts/polish-diff.test.tsscript10.1 KB
scripts/polish-markup.test.tsscript14.0 KB
scripts/polish-partial.test.tsscript2.5 KB
scripts/polish-project.mjsscript5.8 KB
scripts/polish-rules.mjsscript9.8 KB
scripts/rational.mjsscript5.2 KB
scripts/rational.test.tsscript1.9 KB
scripts/raw-decode.mjsscript4.5 KB
scripts/raw-locator.mjsscript2.4 KB
scripts/raw-snapshot.test.tsscript7.3 KB
scripts/references-fences.test.tsscript1.2 KB
scripts/repair-tracker.mjsscript7.5 KB
scripts/repair-tracker.test.tsscript10.8 KB
scripts/report-tools-2.test.tsscript16.1 KB
scripts/report-tools-3.test.tsscript9.3 KB
scripts/report-tools-4.test.tsscript10.1 KB
scripts/report-tools-5.test.tsscript14.6 KB
scripts/report-tools-commands.mjsscript12.9 KB
scripts/report-tools.mjsscript2.6 KB
scripts/report-tools.test.tsscript12.1 KB
scripts/sentence-split.mjsscript3.0 KB
scripts/sentence-split.test.tsscript4.5 KB
scripts/snapshot-readability.test.tsscript7.0 KB
scripts/source-commands.mjsscript12.4 KB
scripts/source-integrity.test.tsscript18.1 KB
scripts/source-refresh.mjsscript5.2 KB
scripts/source-snapshot.mjsscript9.9 KB
scripts/source-snapshot.test.tsscript6.3 KB
scripts/source-view.mjsscript5.0 KB
scripts/sources-citation.mjsscript12.1 KB
scripts/sources-citation.test.tsscript14.8 KB
scripts/tests/contracts-support.tsscript3.5 KB
scripts/tests/delivery-barrier-session.tsscript3.3 KB
scripts/tests/derive-clock.tsscript341 B
scripts/tests/fact-derive-support.tsscript1.9 KB
scripts/tests/fact-unit-dimensions-support.tsscript1.5 KB
scripts/tests/fact-unit-rules-support.tsscript16.8 KB
scripts/tests/fidelity-values-support.tsscript1.6 KB
scripts/tests/fixtures/assets/figure-1.svgother141 B
scripts/tests/fixtures/bad-report.htmlother2.1 KB
scripts/tests/fixtures/boxes.jsonconfig3.2 KB
scripts/tests/fixtures/conversion-probe.htmlother485 B
scripts/tests/fixtures/design-spec.mdmarkdown458 B
scripts/tests/fixtures/fidelity-brief-locked.mdmarkdown571 B
scripts/tests/fixtures/fidelity-brief-pending.mdmarkdown574 B
scripts/tests/fixtures/fidelity/assets/share.csvother40 B
scripts/tests/fixtures/fidelity/claim-graph.mdmarkdown449 B
scripts/tests/fixtures/fidelity/gen-legacy-dev.mjsscript5.0 KB
scripts/tests/fixtures/fidelity/ledger-basic.jsonconfig5.1 KB
scripts/tests/fixtures/fidelity/ledger-legacy-dev.jsonconfig18.3 KB
scripts/tests/fixtures/fidelity/ledger-verify.jsonconfig9.1 KB
scripts/tests/fixtures/fidelity/polish-edits.jsonconfig303 B
scripts/tests/fixtures/fidelity/report-bad.htmlother821 B
scripts/tests/fixtures/fidelity/report-good.htmlother2.4 KB
scripts/tests/fixtures/fidelity/sources/S4.htmlother680 B
scripts/tests/fixtures/fidelity/sources/S4/acronyms.jsonconfig36 B
scripts/tests/fixtures/fidelity/sources/S4/meta.jsonconfig256 B
scripts/tests/fixtures/fidelity/sources/S4/sentences.jsonlother623 B
scripts/tests/fixtures/fidelity/sources/S4/text.txtmarkdown300 B
scripts/tests/fixtures/fidelity/sources/S5/acronyms.jsonconfig3 B
scripts/tests/fixtures/fidelity/sources/S5/meta.jsonconfig235 B
scripts/tests/fixtures/fidelity/sources/S5/sentences.jsonlother53 B
scripts/tests/fixtures/fidelity/sources/S5/text.txtmarkdown10 B
scripts/tests/fixtures/fidelity/sources/S9.txtmarkdown128 B
scripts/tests/fixtures/good-report.htmlother2.7 KB
scripts/tests/fixtures/reference-report.htmlother2.5 KB

references/research-collection.md

View on GitHub
## Phase 0 — Decompose and open the journal

Before spawning anything, decompose the query. Start from "what must be true if the user's intent/spec is true?", not "what looks broken?" Seed `intent-diff.md` with those expected truths before treating code, current docs, or web results as the source of truth:

```
<analysis>
Core question: <the actual information need>
Axes (3+ orthogonal): <axis — what to search, where, why> ...
Codebase relevant: <yes/no> · External: <yes/no> · Collection: <search-only | browser> · Verification likely: <yes/no>
Scale: <axis count, source territories, target document length> · Precision demand: <what a wrong claim costs here> → lifecycle: <single team | research team then refinement team>
</analysis>
```

`Collection` is `search-only` only when a result snippet settles every axis (a date, a version, a spec value); anything that needs reading a source is `browser`.

Then create the session directory:

```bash
mkdir -p .omo/ulw-research/$(date +%Y%m%d-%H%M%S)
```

This is `$SESSION_DIR`. The orchestrator owns the journal: you write every file in it; workers never do (a mass-research dag node writes only its own `waves/<run-key>/<node-id>.md` handoff file, which carries its EXPAND tail and claim candidates; you journal from it). Maintain:

- `brief.md` — the analysis block, the axis list, the expected truths, and the `## Deliverable` record from the interview below (lane, state, formats, each field with `answered_by`).
- `sources-ledger.md` — one line per source the moment it is read: `[S<n>] <url> — <what it is>`. The closing briefing counts sources and domains from this file.
- `sources/` and `fact-ledger.json` — one directory per source snapshot and the ledger built from them. The snapshot and locator contract (raw bytes, converted-only sources, byte locators, `source refresh`) is defined once, in [fact-ledger.md](fact-ledger.md#snapshot-and-locator-contract); every number and quotation reaches the report through a ledger placeholder.
- `wave-<N>-<kind>-<axis>.md` — your digest of each worker return: key findings, sources with URLs, and the worker's EXPAND markers verbatim.
- `expansion-log.md` — per wave: workers spawned, markers gained, leads opened and closed.
- `excursion-log.md` — one ENTER row and one EXIT row per excursion: `excursion_id`, parent claim or axis, ENTER trigger, depth, workers spent, the EXIT rule that closed it, what it changed in the top-level answer (`none` is a valid, required answer), and the ulw-loop steer/evidence id it was mirrored into.
- `intent-diff.md` — orchestrator-owned expected-truth ledger comparing intent/spec/history to observed reality.
- `claim-graph.md` — orchestrator-owned claim graph linking every final assertion to observations, counterevidence, dependencies, and verdict.
- `observation-manifest.md` — orchestrator-owned observation manifest with `observed_at`, temporal validity, artifact paths, and contamination notes.
- `verification-economics.md` — proof-cost ledger mapping claim risk to verification path, deferral decisions, and residual risk.
- `cause-disappearance.md` — cause ledger tracking expected truth, previous observation, `last_seen`, disconfirming observation, and whether the violation is no longer observed.
- `verify-<slug>.md`, `SYNTHESIS.md`, `REPORT.*` from later phases.

Append each digest the moment its worker returns, not in a batch at the end — the journal is your recovery point after context loss and the user's audit trail.

### Run it as a loop, and journal in real time

ulw-loop is ON by default for this mode: when the `ulw-loop` skill is available, register the research axes as loop goals so the run has durable state and survives a compaction. The run's start clock is `outcome.json` `startedAt`, which `outcome init` records once: from `--started-at <ISO>` when you pass it, else from this directory's timestamp name — the closing briefing is computed from it, so create the directory once and never rename it. From that point every finding, source, quote, number, and lead is written into `$SESSION_DIR` **the instant it lands** — never held in the conversation for an end-of-run dump. After any context loss, re-read the brief and the journal before doing anything else, then resume from the open wave.

### Deliverable lane, format, and the empty-info interview

Never block collection on the shape of the deliverable, and never guess it either. Read [references/deliverable-phase.md](deliverable-phase.md) sections 1-6 before writing the brief, then:

1. **Derive first.** Name the lane (`template-strict`, `template-vibe`, `no-format`, `edit-existing`) and the promised formats from the request and its destination (the reference's section 3). When the request refers to an existing deliverable, decide its state with `node "$SKILL_DIR/scripts/report-tools.mjs" outcome state --deliverable <path> --session-dir "$SESSION_DIR"`: `partial` resumes the skeleton on disk, `complete` means edit-existing and never a regenerated report. `$SKILL_DIR` is this skill's own directory, the folder containing this SKILL.md.
2. **Read the requester's format memory** with the memory tool when the harness exposes one: the pointer `system/human/report-style.md`, then `reference/human-report-style.md`; take the choice recorded for the most similar context (same destination kind and audience) as the first option.
3. **Ask only what is still missing**: at most the reference's three questions (destination and format, audience and length, template lineage) in one call to the harness question tool (`ask_user_question` or `request_user_input`), non-blocking where the harness supports it, each with its default first, a free-text "describe the format" path, and "don't care, you decide". Collection starts without waiting; a late answer is folded in until the assembly worker starts, and after that it becomes a re-render request.
4. **Settle source fidelity in the same step.** Seven fields under `## Fidelity` in `brief.md` say what the report must stay faithful to; the reference's section 4 lists each field's values and default. Infer every field before asking anything; an explicit word in the request decides a field (`answered_by: request`), the destination, the sources, and the Phase 0 axes supply the rest (`answered_by: default`):
   - `as of`: a version, release, tag, commit, date, quarter, "as of", "current", or "latest" in the request decides it; a pointed-at repository anchors on its HEAD or tag at run start, pointed-at documents on their own revisions, and a comparison of two anchors records `multiple: <a>, <b>; rows tagged`. Default `latest at run start, <run date>`, the date of the `$SESSION_DIR` timestamp. Ask when the topic is version-bound or time-sensitive (software, APIs, standards, prices, rates, rankings, laws, programs, schedules) and nothing names an anchor; a stable topic takes the default silently.
   - `meaning check`: the QA tier decides (full: `on`, light: `off`); a chat answer on a legal, medical, financial, procurement, or public-facing question with no speed word is `on`. Ask when two explicit signals disagree: a document deliverable under a literal "quick" or "fast", or a light-tier answer that carries both a speed word and a precision demand; `on` goes first.
   - `source floor`: "official docs only", "peer-reviewed only", "SEC filings" give `primary only` or `classes: <host suffixes>`; "from these only" gives `pointed only: <source ids, URLs, or paths>`; pointed-at sources with no restricting word are seeds, not a floor. Default `standard`. Ask when a restricting phrase sits beside an open-ended question whose answer lives outside that floor. Declare each source's provenance when you snapshot it (`source snapshot --provenance primary|secondary`, by the Phase 3b primary-source rule); `primary only` refuses every source not declared `primary`.
   - `derived numbers`: "compare", "growth", "share", "estimate" give `allowed`; "only the sources' own numbers", "no estimates" give `none`. Default `allowed`. Ask when both kinds of phrase share one request. Record only `allowed` or `none` (any other value stops `fact fill` and `fact verify` with a usage error); a partial or free-text answer records `none`.
   - `languages` (never asked): the report in the user's language or the one they named; sources `en`, plus the user's language for a local topic, plus the pointed-at documents' languages; quotes `original` when every source language equals the report language, else `original with translation beside`.
   - `identifier classes` (never asked): from the Phase 0 axes' domain (software: version strings, package names, commit ids; finance: tickers, filing ids, fiscal periods; law: statute, article, and case numbers; science: gene names, DOIs, trial ids), else `none`. It is recorded only; no gate enforces it yet.
   - `fallback` (never asked): `block-quote`, or `omit` on "drop anything you cannot verify"; no other value is accepted.

   Ask nothing when the request, the destination, and the sources decide every field. Otherwise ask only the fields whose trigger fired, in the same question-tool call as the format questions (`ask_user_question` or `request_user_input`), non-blocking where the harness supports it, each with its default first, a free-text path, and "don't care, you decide" (it resolves to the first option, recorded `answered_by: user`). At most four questions ride in that call: `pending` fidelity fields first, then format Q1, Q2, Q3, then `as of`, `meaning check`, `source floor`, `derived numbers`; a question that loses its slot runs on its default and is named, with that default, in the reply that opens collection. When each reading of a field would make part of the request undeliverable, record it `pending` with no default. Collection starts regardless: `source snapshot`, `fact add`, `fact quote`, `fact derive`, `fact retract`, and `fact export` keep working while answers are pending; only the six reader-facing commands (`fact fill`, `fact verify`, `fact pairs`, `fact meaning-check`, `fact polish-apply`, `fact polish-diff`) and the assembly lane wait for the lock, and those commands exit 1 with `fidelity_interview_incomplete` while a field is `pending` or missing. When the assembly lane is about to start, a field still awaiting its answer keeps its default; a `pending` field is asked once more, blocking this time.
5. **Record it.** Write `## Deliverable` into `brief.md` (lane, state, formats, destination, audience, template, format description, each with `answered_by: user|default|request`) plus `## Fidelity` (the block below), and open the manifest in the same step: `node "$SKILL_DIR/scripts/report-tools.mjs" outcome init --promised <formats> --lane <lane> --session-dir "$SESSION_DIR"`. When the ask is to recreate, reproduce, or rebuild something visual, add `Reference compare: required` with the source (URL or file) and the moments to compare; the comparison runs on every round of the build, not only at the end.

#### The fidelity questions

Ask them in the user's language, translating these English texts faithfully and keeping the option order; `<run date>` is the `$SESSION_DIR` date.

- **`as of`**, header `As of`: "Which version or date must the facts hold for? Sources are snapshotted and quoted against that anchor."
  1. `Latest as of today, <run date>` - The state published at the run's start; anything older is flagged as stale.
  2. `Other: a named version, release, or date` - Type it (v3.2, 2026-Q3, a commit or tag). Every quote and number is checked against sources valid at that anchor.
  3. `Don't care, you decide` - Resolves to option 1.
- **`meaning check`**, header `Drift check`: "Numbers and quotes are always script-checked against source snapshots. Should every explanatory sentence also be read against its source quote by a separate reader? Slower and more tokens; it catches widened scope and dropped hedges."
  1. `On` - Every sentence is paired with its source and read; confirmed drift is rewritten from the quote or listed in the briefing.
  2. `Off (faster)` - Script checks only: numbers, units, quotes, qualifiers, bindings. A paraphrase that shifts meaning is not read.
  3. `Other: where to focus` - Type it ("the financial sections", "anything about the competitor"); those pairs go to the strongest reader.
  4. `Don't care, you decide` - Resolves to option 1.
- **`source floor`**, header `Sources`: "Which sources may back a statement of fact?"
  1. `Standard: primary source required for high-risk claims` - Numeric, legal, financial, dated, and causal claims need a primary source plus two independent domains; other claims may cite attributed secondary sources.
  2. `Primary only` - Every fact traces to a first-party source (standard, filing, dataset, official doc); commentary appears only as attributed opinion.
  3. `Only the sources I pointed at` - Ledger rows come from the named files and URLs only; anything else is context, not evidence.
  4. `Other: name the source classes` - Type it ("peer-reviewed only", "no vendor blogs").
  5. `Don't care, you decide` - Resolves to option 1.
- **`derived numbers`**, header `Derived`: "May the report state numbers no source states (sums, growth rates, shares, unit conversions, estimates)?"
  1. `Yes, computed by script and labeled DERIVED` - Each one names its operation and input rows; inputs are checked for unit, period, and population compatibility; an estimate carries its assumptions.
  2. `No, sourced numbers only` - Every number is a MEASURED ledger value; comparisons are stated in words or shown side by side.
  3. `Other: describe` - Type it ("growth rates yes, estimates no"). A partial allowance is recorded as `none`, and the reply that opens collection says so.
  4. `Don't care, you decide` - Resolves to option 1.

#### The `## Fidelity` block

Write it in the same step as `## Deliverable`, provisional defaults included, one line per field with its `answered_by`, plus an `asked:` line naming the fidelity questions sent:

```text
## Fidelity
as of: latest at run start, 2026-10-23 (answered_by: default)
meaning check: on (answered_by: default)
source floor: standard (answered_by: default)
derived numbers: pending (answered_by: pending)
languages: report en; sources en; quotes original (answered_by: default)
identifier classes: version strings, package names, commit ids (answered_by: default)
fallback: block-quote (answered_by: default)
asked: derived numbers
```

Values carry no parentheses; `pending` goes only with `answered_by: pending`. When an answer lands, rewrite that one line with `answered_by: user` and leave `asked:` as history; a late answer after `fact fill` is a re-render: fill again, then verify.

**Re-entry.** An `edit-existing` session, or any session created before this block existed, has no `## Fidelity`. Before any `fact` command, run the inference step above and write the block with `answered_by: default` on every field. `claims index` on a free-form `claim-graph.md` (no claim table) returns `{claims: [], errors: [...]}`, exits 2, and writes no `claims.json`; run `fact verify` without it, and the gate reports each unbound sentence as `unbound_sentence` (major), which blocks delivery until the sentence is bound.

**Calibration.** `fact meaning-check` appends one line per flagged pair to `--calibration` (default `$SESSION_DIR/fidelity-calibration.jsonl`). After each run, also append that run's new lines to the workspace-level `.omo/ulw-research/fidelity-calibration.jsonl`; promotion of `meaning_drift_confirmed` is decided from that log.

Phase 5 opens by turning the recorded fields into `design-spec.md`.

## Phase 1 — Saturation wave

**When the user asked for MASS research, the wave is sized by the topic's angles, not by the roster ceiling.** "mass ulw research", "mulw research", "ulw mass research" — in any language — order over-collection that a team of 8 cannot produce. Where the harness has a dependency-graph surface (the `mass-ulw` skill and its `references/planning.md`), run collection through it instead: one harvest node per angle the topic has, routed across the whole difficulty ladder — mechanical sweeps and per-item harvest batches on the cheapest tier, judgment-shaped angles a step up, cross-territory angles a step above that, and the deepest tier reserved for genuinely hairy contradictions. Each harvest writes its handoff file; when a cluster of angles settles, its open leads re-enter as new nodes until a round finds nothing new, the claims the synthesis rests on are verified per cluster, and the synthesis reduces in ONE architect-tier reducer, fed by per-cluster architect-tier slice reducers when one context cannot hold the result (the strongest reasoning tier substitutes wherever no architect tier exists). Absent that surface, run the same angles with background workers, feeding each worker's open leads into the next round the moment it reports. Everything else in this skill still binds: the deliverable interview, the journal, the claim graph, the convergence rules, and the delivery gates.

Otherwise launch the entire first wave in one turn — every axis at once, as team members if you formed a team, else as background workers. Sequential launches and "start with one and see" defeat the mode.

Scaling floor — more angles always justify more workers:

| Query scope | explore | web | repo-dive | floor |
|---|---|---|---|---|
| Single topic, codebase only | 3 | 0 | 0 | 3 |
| Single topic, web only | 0 | 5 | 1 | 6 |
| Single topic, both | 2 | 4 | 1 | 7 |
| Multi-faceted | 4 | 8 | 2 | 14 |
| Full due diligence | 4 | 9 | 2 | 15 |

**Disambiguate before you expand.** When the topic names something that could resolve several ways — a product, a person, a codename, a version — the first wave settles WHICH entity before any worker researches its history, benchmarks, or controversies: canonical name, first-party URL or account, whether it exists in the claimed category, and a confidence line. An unresolved entity never becomes a premise in a later wave's spawn message; that is exactly how a run starts inventing facts about something that does not exist.

Role protocols — embed the relevant one in each spawn message; every worker gets a unique angle:

- **Codebase (explore), 2-4 workers.** Grep with 3+ keyword variations; structural/AST search; LSP definitions and references; file-name globs; `git log --all -S '<keyword>'` and `--grep` for history including deleted code. Cross-validate hits across tools. Report absolute file paths, patterns with `file:line`, and how findings connect.
- **Web, 3-9 workers, every one loaded with the `browser` skill.** Search with 10+ websearch queries per worker, each with a different operator or angle (see Search craft); a result is a candidate, not a source. Read every candidate in the user's browser through the `browser` skill's attached engine: one named session per worker, one tab per source, several tabs open side by side while sources are compared — `session.tabCreate` / `tabSelect` / `tabClose` where the harness has a JS eval kernel, `bsk tab create --session <session> --url <url>` / `tab select --session <session> <id>` / `tab close --session <session> <id>` / `tab list --session <session>` on the CLI — the page scrolled to its end, a full-page screenshot of each top source saved as provenance, the tab closed once reported. A browser on another machine is reached through the `browser` skill's remote reference or a user-level skill that bridges to it, never replaced by a local headless one. `webfetch` is for raw files only: PDF, JSON, feeds, sitemaps. A host the browser cannot open (WAF, bot score, 403) goes to `ultimate-browsing`, whose provenance travels with the claim: a `snapshot` result cites its `snapshot_timestamp` and is never stated as the live page; a `proxy` result is `untrusted` until a second independent route agrees. Context7 with 3+ queries per known library; grep.app and `gh search code|repos|issues` for real-world usage; official docs via sitemap discovery (`<base>/sitemap.xml`), then targeted pages. A worker whose sources are all fetched text has searched, not read.
- **Repo deep-dive (librarian), 0-2 workers.** Shallow-clone the most relevant repos to `${TMPDIR:-/tmp}`, pin the HEAD SHA, read core modules, follow call chains, return SHA-pinned permalinks.

Example spawn (codebase axis; web and repo-dive follow the same contract with their own protocol):

```
task(subagent_type="explore", run_in_background=true, prompt="TASK: act as a codebase researcher. AXIS: <specific angle>.
This is an explicit exhaustive-research assignment. Your default retrieval budget and stop-when-answered rules do not apply — run the full protocol below and report every lead.
SCOPE: find everything in this codebase related to <angle>: <what complete looks like>.
PROTOCOL: grep 3+ keyword variations; structural search; LSP references; globs; git history (-S and --grep). Cross-validate across tools. Report absolute paths and file:line patterns.
Tag every claim OBSERVED: <URL | file:line | command>. End your reply with the ## EXPAND tail: '- LEAD: <discovery> — WHY: <why> — ANGLE: <search>' per lead, '- DEAD END: <lead> — PROBED: <what exhausted it>', or 'none — <reason>'.")
```

## Phase 2 — Expand until convergence

This loop is what makes the mode research rather than search. Collect returns as they land — and in team mode, act on each lead the moment a member raises it, never waiting for the full wave or a member's final reply:

1. Journal the return the moment it lands, never at the end of the wave: digest plus verbatim EXPAND markers into `wave-<N>-<kind>-<axis>.md`, appending each new source, quote, and number to the observation manifest in the same beat — after a compaction the journal, not your memory, is the state. Snapshot each source the moment it is read (`node "$SKILL_DIR/scripts/report-tools.mjs" source snapshot --id S<n> --url <url> --session-dir "$SESSION_DIR"`; a page read in the browser is saved and imported with `--html <file>`, recorded as lower-trust `model` extraction), then register every number and quote you will use from that snapshot: `fact add --source S<n> --find "<anchor text>" --label <label>` (or `--table <id> --row <label> --col <label>`, or `--page <n> --line <n>`) and `fact quote --source S<n> --match "<words>" --label <label>`. Read the printed row and its 60 characters of source context before using it; a wrong row is retracted (`fact retract F<n> --reason <why>`), never edited. The observation's quote or line anchor in the manifest is that fact id.
2. Deduplicate new markers against `expansion-log.md` — every lead ever seen, not just confirmed ones, or rejected leads resurface each wave.
3. Spawn an expansion worker immediately for each new unchecked lead:

```
task(category="deep-low", load_skills=["browser"], run_in_background=true, prompt="TASK: expansion wave <N> — investigate: <lead>.
PARENT: <which return surfaced it>. This is an explicit exhaustive-research assignment; budgets do not apply.
<role protocol for the lead's territory — web protocol for external leads, explore protocol for codebase leads>
Tag every claim OBSERVED: <URL | file:line | command>. End your reply with the ## EXPAND tail.")
```

### Excursions — dive deep on a new find, then surface back out

The Phase 0 core question is the fixed goal of the run and never drifts. An excursion is a BOUNDED detour off the wave plan to chase something a return surfaced — you go deep, settle it, and come back up to the question you were hired to answer.

**ENTER (dive) only on a trigger.** One of these must hold, and you name which one:

1. The find contradicts a claim already locked in `claim-graph.md`.
2. It would change the final answer or a recommendation if it turned out to be true.
3. It exposes a source territory no axis owns, so nobody else will ever reach it.
4. The user's steering points at it — their words are the trigger, quoted verbatim.

Interest alone is not a trigger. Anything without one stays a queued lead in `expansion-log.md`, and the wave plan continues.

**Budget the dive before you take it.** State the worker count and the probe count for this level in the ENTER row. An excursion may spawn at most ONE nested sub-excursion; a third level means the thing has become its own research question — surface immediately and either promote it to a real axis with its own worker or record it as an out-of-scope gap in `SYNTHESIS.md`. When the `ulw-loop` skill is driving the run, a promotion is `omo-agent-toolkit ulw-loop steer --kind add_subgoal --title "<axis>" --objective "<what it must answer>" --evidence "<what surfaced it>" --rationale "<why the plan changes>"`.

**EXIT (surface) the moment any of these holds** — you do not need all of them:

- The ENTER trigger is resolved: the claim is confirmed, refuted, or its dependency is closed.
- Two consecutive probes changed nothing in the parent answer.
- The finding stops moving any claim's status — diminishing return is an exit, not a reason to push harder.
- The level's stated budget is spent.

**Fold back on the way out.** Every EXIT writes one line saying what the excursion changed in the top-level answer, and `none — <reason>` is a legitimate, required outcome; an excursion whose result is silently dropped is a lost run. Update the parent claim node or axis digest with the result, then mirror the whole excursion into the loop ledger — `omo-agent-toolkit ulw-loop steer --kind annotate_ledger --evidence "<what the excursion observed>" --rationale "<what it changed, or none>"`, and `omo-agent-toolkit ulw-loop record-evidence --goal-id <id> --criterion-id <id> --status pass|fail|blocked --evidence "<artifact>"` when it settled a criterion. After a compaction, `omo-agent-toolkit ulw-loop status --json` plus `excursion-log.md` tell you which excursions are still open.

**Anti-drift.** After every EXIT, re-read the core question in the journal and confirm the run still answers it. Three consecutive excursions that changed nothing end excursions for the run: converge on what you have.

4. Record the wave in `expansion-log.md`: spawned, markers gained, leads opened/closed.
5. **Relay the user's steering to everyone.** When the user changes scope, cadence, target sources, language, or format mid-run, broadcast it to every live worker and member immediately and record the exact wording in `expansion-log.md`. Steering only you saw silently splits the swarm's assignment from the user's actual ask.

**Convergence — the only stop rules while this mode is active.** Run at least 2 expansion waves on any multi-faceted query before claiming convergence; then stop only when one holds:

- Zero unchecked leads remain — each investigated or closed as duplicate/dead end.
- 3 consecutive waves produced no new actionable leads.
- Expansion depth reached 5 waves — pause, show the open leads, and ask the user whether to extend.

**Never end the run on a worker's completion.** Workers finishing is not the deliverable; your synthesis is. Reserve the last fifth of the run's context and time for Phases 4-5 and stop opening waves the moment that reserve is threatened. A converged answer with two open leads beats nine finished workers and no report.
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 322kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 322kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 322kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 198kUpdated Claude CodeCodexVerifiedInstall