You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Wire the knowledge plugin's library_diruserConfig seam into the youtube pipeline at runtime. Before this, library_dir was configurable and persisted by setup, but the youtube pipeline never consumed it — a non-default value (e.g. docs/knowledge) did not redirect watch/transcript/queue artifacts; they still landed at the consuming repo root via resolveWorkRoot()'s CLAUDE_PROJECT_DIR fallback.
How
run.mjs learns an optional leading --work-root <dir> flag and translates it into the YOUTUBE_WORK_ROOT env var the extraction scripts' resolveWorkRoot() already reads (forwarded to the spawned child). Arg parsing + child-env layering extracted to lib/run-args.js (pure, unit-tested in lib/run-args.test.js).
youtube/SKILL.md — new "Artifact landing (work root)" section: when ${user_config.library_dir} is non-default, pass --work-root "${CLAUDE_PROJECT_DIR}/${user_config.library_dir}" on everyrun.mjs invocation (call sites enumerated). Default/empty/unexpanded → no flag; resolveWorkRoot() fallback intact.
setup/SKILL.md Output — restores the stronger "library_dir governs where youtube artifacts land" wording that was softened while the seam was unwired; book-distill remains the documented exception.
plugin.json0.3.0 → 0.4.0 + CHANGELOG entry.
Decision (unbriefed, called out)
Decision
What it changes
Basis (evidence)
Cross-platform CLI arg (run.mjs --work-root) instead of the issue's literal "export YOUTUBE_WORK_ROOT before invoking" as a shell env prefix
Adds a small optional flag to run.mjs (issue framed the change as SKILL.md-only)
An inline YOUTUBE_WORK_ROOT=… node prefix is bash-only and fails under PowerShell — run.mjs's own docstring already avoids NODE_PATH=… node for exactly this reason (Cross-Platform-by-Default). The in-repo idiom for wiring a ${user_config.X} seam into a command is a double-quoted CLI arg (claude-troubleshooting passes --data-dir "…"). resolveWorkRoot() is left untouched per the issue.
Verification
vitest run — 221/221 pass (incl. new lib/run-args.test.js)
tsc --noEmit — clean
claude plugin validate + scripts/validate-plugins.sh — pass
scripts/run-plugin-tests.sh — 18 checks + all plugin tests pass
e2e smoke: run.mjs --work-root X <probe> → child sees YOUTUBE_WORK_ROOT=X; no flag → falls back to CLAUDE_PROJECT_DIR; bad flag → clean usage exit 2
Refs melodic-software/medley#1466
Note
Low Risk
Additive launcher flag and skill documentation; default artifact paths unchanged when library_dir is ..
Overview /knowledge:youtube now respects library_dir. Non-default values redirect watch, transcript, and queue .work/… output under the configured directory instead of always using the repo root.
The skill passes a leading run.mjs --work-root <dir> on every extraction invocation; lib/run-args.js parses that flag and sets YOUTUBE_WORK_ROOT for the child so existing resolveWorkRoot() behavior applies. Default . omits the flag (unchanged fallback). Skill docs require agent-written queue/claim/output paths to use the same root so scripts and hand-edited artifacts stay aligned.
setup output again states that library_dir governs YouTube artifact landing (book-distill still documented as the exception). Plugin version 0.4.0 + CHANGELOG.
Reviewed by Cursor Bugbot for commit eb2138d. Bugbot is set up for automated code reviews on this repo. Configure here.
…into the pipeline
The youtube pipeline defined a `library_dir` userConfig seam but never
consumed it: a non-default value (e.g. docs/knowledge) did not redirect
watch/transcript/queue artifacts, which still landed at the repo root via
resolveWorkRoot()'s CLAUDE_PROJECT_DIR fallback.
Wire it end to end:
- run.mjs learns an optional leading `--work-root <dir>` flag and translates
it into the YOUTUBE_WORK_ROOT env var the extraction scripts already read.
A double-quoted CLI arg is cross-platform; an inline `YOUTUBE_WORK_ROOT=… node`
prefix is bash-only (the same reason run.mjs already avoids `NODE_PATH=… node`).
Arg parsing + child-env layering extracted to lib/run-args.js (unit-tested).
- youtube/SKILL.md documents passing `--work-root "${CLAUDE_PROJECT_DIR}/${user_config.library_dir}"`
on every run.mjs invocation when library_dir is non-default; unchanged (no flag)
for the repo-root default, keeping the resolveWorkRoot() fallback intact.
- setup/SKILL.md Output restores the stronger "library_dir governs where youtube
artifacts land" wording softened while the seam was unwired (book-distill remains
the documented exception).
- knowledge plugin.json 0.3.0 -> 0.4.0 + CHANGELOG entry.
Refs melodic-software/medley#1466
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk: medium. Cursor Bugbot did not complete (check skipped due to usage limit), so the required automated review signal is incomplete. Human review is needed before merge; no reviewers were assigned because the only prior human contributor is the PR author.
Sent by Cursor Approval Agent: Pull Request Router and Approver
…ved work root
Addresses a Codex review finding: the `--work-root` flag redirected only the
script-written artifacts under a non-default `library_dir`, while the skill's
hand-materialized `.work/<watch-epic>/…` paths (QUEUE.md, claim stubs, and the
agent-authored Output-contract deliverables) still resolved to the repo root —
splitting the queue table from its concurrency claims across two directories.
Document that every `.work/<watch-epic>/…` path in the skill and its context
files is relative to the same resolved work root, so a non-default library_dir
co-locates the whole slice instead of splitting it.
Good catch — fixed in eb2138d. The --work-root flag only redirected the script-written artifacts; the skill's hand-materialized .work/<watch-epic>/… paths (QUEUE.md, claim stubs, agent Output-contract deliverables) still resolved to the repo root, splitting the table from its claims under a non-default library_dir. The SKILL now documents that every.work/<watch-epic>/… path — queue materialization, claims, and agent-authored artifacts alike — is relative to the same resolved work root, so the whole slice co-locates.
The reason will be displayed to describe this comment to others. Learn more.
Risk: medium. Cursor Bugbot did not complete (check skipped due to usage limit), so the required automated review signal is incomplete. Human review is needed before merge; no reviewers were assigned because the only prior contributor to these paths is the PR author.
Sent by Cursor Approval Agent: Pull Request Router and Approver
…ne (#146)
Retrofit the `course-digest` skill (Dometrain/Teachable course
extraction and repo-applicability synthesis) into the `knowledge`
plugin, replicating the shipped `youtube` vendoring precedent. Pre-baked
decisions from the source issue were honored, not re-litigated.
## What changed
- **Vendored shared libs** — `@melodic/repo-analysis` +
`@melodic/video-digestion` copied (runtime source only) into
`extraction/vendor/`, each with its own trimmed `package.json` (real
`name`/`version`/`exports`; `video-digestion` keeps its `imghash`
runtime dep). `package.json` deps rewritten to `file:./vendor/*`.
Byte-identical to the youtube skill's vendored copies (dedup of the two
is tracked separately).
- **Launcher trio** — `run.mjs` / `register-hook.mjs` /
`resolve-hook.mjs` resolve bundled deps from `${CLAUDE_PLUGIN_DATA}` via
an ESM resolve hook. `setup-deps.mjs` installs node deps into
`${CLAUDE_PLUGIN_DATA}` (sha256 fingerprint over `package.json` +
`vendor/`, stamped `.course-extraction.stamp`) **and provisions
Playwright's Chromium** into `${CLAUDE_PLUGIN_DATA}/ms-playwright`.
`run.mjs` pins `PLAYWRIGHT_BROWSERS_PATH` to that path so install-path
and runtime-lookup-path stay in lockstep.
- **Credentials (adapter-native, no sensitive userConfig)** — login via
the user's own `COURSE_*`/`TEACHABLE_*` shell env vars (prefix driven by
`platformConfig.authEnvPrefix`), interactive manual-login fallback.
Session cookies re-homed out of the consumer repo to
`${CLAUDE_PLUGIN_DATA}/auth/<platform>.auth-state.json` via
`lib/auth-store.js`.
- **Decoupling** — inlined `tsconfig.json` (no cross-repo `extends`);
genericized `SKILL.md` + `context/`/`reference/` paths to
`${CLAUDE_PLUGIN_ROOT}`; routed script invocations through `run.mjs`;
namespaced self-referencing slashes to `/knowledge:course-digest` (and
cross-refs to `/knowledge:youtube`); rewrote `!`-precompute probes to
cross-platform `node -e`; fixed the `build-course-json.js` JSDoc example
path.
- **Version + catalog** — `knowledge` plugin `0.4.0` → `0.5.0` (rebased
on #145, which took `0.4.0`), CHANGELOG `0.4.0` entry added (plus a
backfilled `0.3.0` entry that PR #138 had omitted). Extended plugin
`keywords` and marketplace `tags` with course/video terms.
- Excluded the ~2 GB `data/` tree (consumer machine state, per the
issue).
## Verification
- `claude plugin validate --strict` — clean (plugin + marketplace).
- `npx vitest run` — **10 files, 91 tests pass**; `tsc --noEmit` clean;
markdownlint clean.
- **Clean-repo smoke** (scratchpad, `CLAUDE_PLUGIN_DATA`=temp, **local
`node_modules` absent**):
- `setup-deps.mjs` installed deps + Chromium into the data dir; re-run
early-exits (idempotent).
- Vendored `@melodic/*` + `playwright` bare imports resolved **from
`${CLAUDE_PLUGIN_DATA}`** (proved by running a pipeline script with no
local `node_modules`).
- Dry run reached the **auth branch**: Chromium launched from the
data-dir path and control entered `adapter.authenticate` → `clerkLogin`,
failing exactly as expected on a fake login page (no real credentials).
## Implementation decisions (reviewer visibility)
- **Auth is keyed per-platform**, not per-course. The source issue
framed the re-home as "location, not behavior"; per-platform is a slight
behavior change but the correct shape (a login session is a platform
session, reused across courses).
- **Dev/test fallback** when `CLAUDE_PLUGIN_DATA` is unset:
`${home}/.claude/course-digest/auth/…` — machine-local, persistent, out
of the consumer repo.
- `--course-dir` stays caller-provided; full `library_dir` → course-dir
auto-resolution (the youtube work-root analog) is **not** wired here —
possible follow-up.
- **Manual-login-under-headless** may not function (documented as a
known limitation, not a blocker); env-var + cookie-reuse carry the
skill.
Refs melodic-software/medley#1409
<!-- CURSOR_SUMMARY -->
---
> [!NOTE]
> **Medium Risk**
> Adds Playwright automation, course-platform credentials, and large new
surface area (browser auth, vendored deps, long-running extraction), but
credentials stay in env/plugin data and behavior is documented/tested
rather than changing existing skills.
>
> **Overview**
> **Knowledge plugin 0.5.0** adds **`/knowledge:course-digest`**: a
multi-phase workflow to discover, extract, and synthesize **Dometrain**
and **Teachable** courses into repo-applicable artifacts
(`repo-candidates.md`, `action-items.md`), with single YouTube videos
still routed to `/knowledge:youtube`.
>
> The PR ships a **bundled Node extraction stack** (Playwright
orchestration, ffmpeg/ImageMagick frame tooling, companion-repo
analysis) following the **youtube** pattern: `@melodic/repo-analysis`
and `@melodic/video-digestion` vendored under `extraction/vendor/`,
**`setup-deps.mjs`** installing into `${CLAUDE_PLUGIN_DATA}` plus
Chromium under `ms-playwright`, and **`run.mjs`** resolving deps and
pinning `PLAYWRIGHT_BROWSERS_PATH`. **Platform adapters** (`dometrain`,
`teachable`) implement a shared contract (transcript, HLS, resources,
URLs); shared **Clerk** / **Teachable SSO** auth and **Mux** /
**Hotmart** player modules live under `lib/`.
>
> **Auth** uses shell `COURSE_*` / `TEACHABLE_*` (from
`platformConfig.authEnvPrefix`) with manual-login fallback; session
state is stored per platform at
**`${CLAUDE_PLUGIN_DATA}/auth/<platform>.auth-state.json`**, not in the
consumer repo. Skill docs cover the 8-phase pipeline, storage schema,
evals, and prerequisites; marketplace/plugin metadata and CHANGELOG are
updated (including a backfilled **0.3.0** entry).
>
> <sup>Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit
477e254. Bugbot is set up for automated
code reviews on this repo. Configure
[here](https://www.cursor.com/dashboard/bugbot).</sup>
<!-- /CURSOR_SUMMARY -->
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Wire the knowledge plugin's
library_diruserConfigseam into the youtube pipeline at runtime. Before this,library_dirwas configurable and persisted bysetup, but the youtube pipeline never consumed it — a non-default value (e.g.docs/knowledge) did not redirect watch/transcript/queue artifacts; they still landed at the consuming repo root viaresolveWorkRoot()'sCLAUDE_PROJECT_DIRfallback.How
run.mjslearns an optional leading--work-root <dir>flag and translates it into theYOUTUBE_WORK_ROOTenv var the extraction scripts'resolveWorkRoot()already reads (forwarded to the spawned child). Arg parsing + child-env layering extracted tolib/run-args.js(pure, unit-tested inlib/run-args.test.js).youtube/SKILL.md— new "Artifact landing (work root)" section: when${user_config.library_dir}is non-default, pass--work-root "${CLAUDE_PROJECT_DIR}/${user_config.library_dir}"on everyrun.mjsinvocation (call sites enumerated). Default/empty/unexpanded → no flag;resolveWorkRoot()fallback intact.setup/SKILL.mdOutput — restores the stronger "library_dirgoverns where youtube artifacts land" wording that was softened while the seam was unwired;book-distillremains the documented exception.plugin.json0.3.0→0.4.0+ CHANGELOG entry.Decision (unbriefed, called out)
run.mjs --work-root) instead of the issue's literal "exportYOUTUBE_WORK_ROOTbefore invoking" as a shell env prefixrun.mjs(issue framed the change as SKILL.md-only)YOUTUBE_WORK_ROOT=… nodeprefix is bash-only and fails under PowerShell —run.mjs's own docstring already avoidsNODE_PATH=… nodefor exactly this reason (Cross-Platform-by-Default). The in-repo idiom for wiring a${user_config.X}seam into a command is a double-quoted CLI arg (claude-troubleshootingpasses--data-dir "…").resolveWorkRoot()is left untouched per the issue.Verification
vitest run— 221/221 pass (incl. newlib/run-args.test.js)tsc --noEmit— cleanclaude plugin validate+scripts/validate-plugins.sh— passscripts/run-plugin-tests.sh— 18 checks + all plugin tests passrun.mjs --work-root X <probe>→ child seesYOUTUBE_WORK_ROOT=X; no flag → falls back toCLAUDE_PROJECT_DIR; bad flag → clean usage exit 2Refs melodic-software/medley#1466
Note
Low Risk
Additive launcher flag and skill documentation; default artifact paths unchanged when library_dir is
..Overview
/knowledge:youtubenow respectslibrary_dir. Non-default values redirect watch, transcript, and queue.work/…output under the configured directory instead of always using the repo root.The skill passes a leading
run.mjs --work-root <dir>on every extraction invocation;lib/run-args.jsparses that flag and setsYOUTUBE_WORK_ROOTfor the child so existingresolveWorkRoot()behavior applies. Default.omits the flag (unchanged fallback). Skill docs require agent-written queue/claim/output paths to use the same root so scripts and hand-edited artifacts stay aligned.setupoutput again states thatlibrary_dirgoverns YouTube artifact landing (book-distillstill documented as the exception). Plugin version 0.4.0 + CHANGELOG.Reviewed by Cursor Bugbot for commit eb2138d. Bugbot is set up for automated code reviews on this repo. Configure here.