@staff0rd/assist 0.338.0 → 0.340.0

package/README.md

@@ -72,8 +72,7 @@ After installation, the `assist` command will be available globally. You can als
 - `/sql` - Query a MSSQL database via assist sql
 - `/verify` - Run all verification commands in parallel
 - `/verify-new` - Add a new verify:\* run command to assist.yml
-- `/transcript-format` - Format meeting transcripts from VTT files
-- `/transcript-summarise` - Summarise transcripts missing summaries
+- `/transcripts` - Format and summarise meeting transcripts end to end
 - `/voice-setup` - Download required voice models (VAD, STT)
 - `/voice-start` - Start the voice interaction daemon
 - `/voice-stop` - Stop the voice interaction daemon
@@ -254,8 +253,8 @@ The first backlog command in a repository that still has a local `.assist/backlo
 - `assist complexity maintainability [pattern]` - Calculate maintainability index per file (`--ignore <glob>`, repeatable, excludes extra files on top of `complexity.ignore`). A file can declare its own threshold with a `// assist-maintainability-override: N` comment in its first ~10 lines (integer 0-100); that value replaces the global `--threshold` for that file only and the output annotates it with `(override: N)`
 - `assist complexity sloc [pattern]` - Count source lines of code per file
 - `assist transcript configure` - Configure transcript directories
-- `assist transcript format` - Convert VTT files to formatted markdown transcripts
-- `assist transcript summarise` - List transcripts that do not have summaries
+- `assist transcript list` - List raw .vtt filenames waiting in the pick-up directory
+- `assist transcript move <file>` - Convert a raw .vtt to a dated markdown transcript and archive the original
 - `assist voice setup` - Download required voice models (VAD, STT)
 - `assist voice start` - Start the voice daemon (always-on, listens for wake word)
 - `assist voice start --foreground` - Start in foreground for debugging

package/allowed.cli-reads

@@ -54,6 +54,7 @@ assist sql columns
 assist sql query
 assist sql tables
 assist status-line
+assist transcript list
 assist verify
 assist voice
 az account list

package/allowed.cli-writes

@@ -17,5 +17,3 @@ assist handover
 assist notify
 assist signal next
 assist sync
-assist transcript format
-assist transcript summarise

package/claude/commands/transcripts.md

@@ -0,0 +1,49 @@
+---
+description: Format and summarise meeting transcripts end to end
+---
+
+Drain the transcript pick-up directory: for every raw `.vtt`, convert it to a dated markdown transcript and write a summary alongside it. Loop until nothing is left.
+
+Repeat the steps below until `assist transcript list` prints nothing.
+
+## Step 1: List
+
+```bash
+assist transcript list
+```
+
+Each line is the bare filename of a raw `.vtt` waiting to be processed.
+
+- **No output** — every transcript is done. Stop.
+- **One or more lines** — take the first filename and process it with the remaining steps, then return here.
+
+## Step 2: Determine the client and date
+
+Work out the **client** and meeting **date** (`YYYY-MM-DD`) for the file:
+
+- Prefer a client name and `YYYY-MM-DD` date already present in the filename.
+- If the filename isn't clear, read the raw `.vtt` to infer them from the meeting content. The file lives at `<vttDir>/<filename>`, where `vttDir` is the `transcript.vttDir` value in `.claude/assist.yml` (or `assist.yml`).
+- Only if you still cannot confidently determine the client or date, ask the user for the missing value(s). Do **not** ask when you can infer them.
+
+## Step 3: Move
+
+Convert the raw `.vtt` to a dated markdown transcript and archive the original:
+
+```bash
+assist transcript move "<filename>" --date <YYYY-MM-DD> --client "<client>"
+```
+
+`move` prints two labelled paths:
+
+- **`Formatted transcript:`** — the formatted transcript it just wrote
+- **`Summary target:`** — the summary target path to create in the next step
+
+## Step 4: Write the summary
+
+Read the formatted transcript (the `Formatted transcript:` path) and write a concise summary to the summary target path (the `Summary target:` path). The summary must start with a link back to the transcript:
+
+```
+[Full Transcript](<relative path from the summary file's directory to the formatted transcript>)
+```
+
+Then return to Step 1.

package/claude/settings.json

@@ -38,8 +38,7 @@
 			"SlashCommand(/devlog)",
 			"SlashCommand(/refactor)",
 			"SlashCommand(/review-comments)",
-			"SlashCommand(/transcript-format)",
-			"SlashCommand(/transcript-summarise)",
+			"SlashCommand(/transcripts)",
 			"SlashCommand(/voice-setup)",
 			"SlashCommand(/voice-start)",
 			"SlashCommand(/voice-stop)",
@@ -52,8 +51,7 @@
 			"Skill(devlog)",
 			"Skill(refactor)",
 			"Skill(review-comments)",
-			"Skill(transcript-format)",
-			"Skill(transcript-summarise)",
+			"Skill(transcripts)",
 			"Skill(journal)",
 			"Skill(standup)",
 			"Skill(voice-setup)",

package/dist/allowed.cli-reads

@@ -54,6 +54,7 @@ assist sql columns
 assist sql query
 assist sql tables
 assist status-line
+assist transcript list
 assist verify
 assist voice
 az account list

package/dist/allowed.cli-writes

@@ -17,5 +17,3 @@ assist handover
 assist notify
 assist signal next
 assist sync
-assist transcript format
-assist transcript summarise