sawz/last30days-skill
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: vs mode N full passes + --competitors auto-discovery + (/Last30Days) title (#312)

Browse Source 
* feat: vs mode runs N full passes; --competitors wraps vs with auto-discovery

Unifies vs-mode and --competitors onto one fanout architecture. A topic
containing "vs" / "versus" now runs N full pipeline.run() calls in parallel
(reverting the one-pass latency optimization that removed per-entity
depth); --competitors becomes a SKILL.md-level shortcut where the hosting
reasoning model (Claude Code, Codex, Hermes, Gemini) discovers N peers via
its own WebSearch, runs Step 0.55 per entity, and invokes the engine with
a vs-topic + --competitors-plan JSON.

Changed:
- vs-mode: N full passes in parallel via fanout (was 1 merged pass).
- --competitors: SKILL.md shortcut for vs-mode-with-discovery. Engine flag
  kept for headless/cron use. LAW 7-style stderr reframed to lead with the
  hosting-model path (use WebSearch + --competitors-plan) instead of
  BRAVE_API_KEY. Footer BRAVE/SERPER nudge suppressed when --plan or
  --competitors-plan present (hosting model already has WebSearch).

Added:
- --competitors-plan JSON flag: per-entity {x_handle, x_related, subreddits,
  github_user, github_repos, context}. Accepts inline JSON or file path.
  subrun_kwargs_for helper is the single source of truth for per-entity
  kwargs — no closure-default fallthrough from main scope.
- Per-entity save files: each entity's sub-run produces its own
  {slug}-raw.md with a single-row Resolved Entities block.
- --polymarket-keywords filter for ambiguous single-token topics.

Fixed:
- test_competitor_subrun_isolation regression suite locks in 3.0.12's
  no-leak invariant (main flags do not inherit into peer sub-runs).
- Updates test_regression.py for the new comparison-mode payload shape.

Bumps plugin.json to 3.0.13. 1,219 tests passing.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix: comparison title attribution — (Last 30 Days) → (/Last30Days)

User feedback on 3.0.13 dogfood runs (Kanye vs Drake, Mercer Island,
Figma): the comparison-mode synthesis title should attribute to the
slash command rather than restate the date range.

Three SKILL.md occurrences updated. Pure documentation change. Bumps to
3.0.14.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Matt Van Horn
2026-04-22 21:31:00 -07:00
committed by GitHub
parent 00d01933e0
commit 949bcf8942
18 changed files with 2022 additions and 71 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CHANGELOG.md
+25
View File
@@ -5,6 +5,31 @@ All notable changes to this project will be documented in this file.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
## [3.0.14] - 2026-04-22
### Changed
- **Comparison-mode title attribution.** The synthesis title for vs-mode and `--competitors` outputs changes from `What the Community Says (Last 30 Days)` to `What the Community Says (/Last30Days)`. Surfaces the slash-command identity instead of restating the date range. Three SKILL.md occurrences updated; pure documentation change.
## [3.0.13] - 2026-04-22
### Changed
- **vs mode runs N full passes in parallel, one per entity.** Architectural revert of the 3-pass → 1-pass latency optimization from an earlier version. `/last30days "OpenAI vs Anthropic vs xAI"` now runs three full `pipeline.run()` calls in parallel via the same fanout `--competitors` uses, producing three `*-raw.md` save files plus a merged comparison output. Each entity gets its own Step 0.55-grade targeting, own primary X handle weight, own subreddit scoping — apples-to-apples depth instead of the one-pool merged retrieval the single-pass path produced. Parallel execution keeps wall clock ≈ single pass.
- **`--competitors` is now a SKILL.md-level shortcut for vs-mode with auto-discovery.** The hosting reasoning model (Claude Code, Codex, Hermes, Gemini, any agent with WebSearch) performs discovery and Step 0.55 per entity via its own WebSearch tool, then invokes the engine with a vs-topic and `--competitors-plan` JSON. The engine flag remains for headless/cron use with BRAVE/EXA/SERPER/PARALLEL/OPENROUTER keys (engine-internal `auto_resolve` stays as fallback).
- **LAW 7-style stderr for `--competitors` with no backend** now leads with the hosting-model path (WebSearch + Step 0.55 + `--competitors-plan`) instead of `BRAVE_API_KEY`. API-key framing moved to a secondary "headless" section.
### Added
- **`--competitors-plan` JSON flag** for per-entity Step 0.55 targeting. Schema: `{entity_name: {x_handle?, x_related?, subreddits?, github_user?, github_repos?, context?}}`. Accepts inline JSON or a file path (matches `--plan`). When present for an entity, skips engine-internal `auto_resolve` and uses the provided values; missing fields fall back to `auto_resolve` (if backend) or planner defaults. Case-insensitive entity matching. The `subrun_kwargs_for` helper is the single source of truth for per-entity kwargs — no closure-default fallthrough from main scope.
- **Per-entity save files** when `--save-dir` is set on a vs-mode or `--competitors` run. Each entity's sub-run produces its own `{slug}-raw.md` with a single-row Resolved Entities block — matches historical vs-mode behavior (N passes → N save files).
- **`--polymarket-keywords "kw1,kw2"`** to filter Polymarket matches for ambiguous single-token topics (e.g., "Warriors" → `nba,gsw,golden-state` kills Glasgow Warriors rugby and Honor of Kings Rogue Warriors noise).
### Fixed
- **BRAVE/SERPER footer nudge suppressed** when `--plan` or `--competitors-plan` is present. The nudge told Claude Code users to set an API key when they already have WebSearch via the hosting model. Nudge still fires for true headless runs (no `--plan`, no backend) where the advice is correct.
- **Override-leak regression testing.** 3.0.12 already fixed the main-topic `--subreddits` / `--x-handle` / `--github-*` from leaking into peer sub-runs via explicit per-entity kwargs scrubbing. This release adds a 4-test regression suite (`test_competitor_subrun_isolation.py`) locking in the invariant.
## [3.0.12] - 2026-04-22
### Fixed