agentsclimarketplace

Debrief

Skill bks-lab/open-bridge/skills/debrief

Your AI coding agent starts every session knowing your repos, your clients, and how you work — a plain git repo of markdown + YAML it reads at session start, independent of model or frontend. Context compounds instead of restarting. MIT.

Install
npx -y skills add bks-lab/open-bridge --skill debrief

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 8 stars8 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Meeting and transcript processing — classifies meeting type, applies name corrections, extracts insights across 7 categories, generates protocols, proposes tasks with project field mapping. Checkpoint-based workflow. Supports full 8-phase flow (with GitHub reconciliation + distribution email) and quick 5-phase flow; `--all` and `--date YYYY-MM-DD` batch flags absorb the former /process-meeting. Trigger: "/debrief", "/debrief {path}", "/debrief --quick", "/debrief --all", "/debrief --date YYYY-MM-DD", "process meeting", "transcript processing", "create minutes", "process-meeting", "meeting minutes", "process transcript".

SKILL.md

5.3 KB, as published. Nobody here has run it

Debrief

Process a meeting transcript. Read the referenced file ONLY when triggered.

Sources and flow variants

Sources scanned: legacy work.imports_dir plus doc_sensor.scan_paths entries with kind: recordings (catches PARA tree 2_AREAS/Import_Audio/ after the PARA migration). Already-archived basenames in work/archive/days/ are skipped as redundant.

Flow variants:

  • Full 8-phase flow (references/full-workflow.md) with GitHub task-reconciliation (references/task-reconciliation.md) and optional distribution email (references/distribution-email.md) to meeting participants.
  • Quick 5-phase flow (references/quick-workflow.md).
  • Batch flags: --all processes every unprocessed transcript in imports/; --date YYYY-MM-DD processes only that day. (Both formerly lived in /process-meeting, now merged here.)

Decision Tree

User wants to...
├── Full processing (8 phases)              → Read references/full-workflow.md
├── Quick processing (5 phases)             → Read references/quick-workflow.md
├── Classify a transcript only              → Read references/classification.md
├── Apply name corrections                  → Read references/classification.md (§ Name Corrections)
├── Generate a protocol from notes          → Read references/protocol-templates.md
├── Match actions to existing issues        → Read references/task-reconciliation.md
│                                             (called from Phase 5)
├── Pull calendar/chat context for transcript → Read references/full-workflow.md § Phase 1.5
│                                             (driven by integrations.meeting_context.*)
├── Send meeting summary email              → Read references/distribution-email.md
│                                             (called from Phase 7, after updates + protocol)
└── Questions about debrief                 → Answer from this file

Phase order (hard rule)

  1. Find → 1.5. Context Lookup (only if any integrations.meeting_context.* block in bridge-config.yaml is enabled with consumers: [debrief]) → 2. Classify + corrections → 3. Extract →
  2. (n/a — merged into 3) → 5. Task Reconciliation + Execute → 6. Protocol → 7. Distribution Email → 8. Work-log + Archive

Updates before protocol (so wiki has real URLs). Protocol before email (so email links to real issue updates AND the written wiki page). Email before archive (so transcript is still next to the workflow).

Flow selector

FlagFlowWhen to use
(none)Full 7-phaseMeetings with stakeholders, decisions, action items. Protocol required.
--quickQuick 5-phaseLoose notes, personal debriefs, no formal protocol needed.
{path}Full or quickPath to transcript file. Flow selection via other flags.

Both flows are checkpoint-gated — user confirms at classification and task creation. Neither flow writes to GitHub, wiki, or archive without explicit [y].

Transcription pipeline (optional)

Debrief consumes transcripts; it does not transcribe. A transcription worker is an optional integration — never a dependency:

  • No worker configured (the default): transcribe with any tool and drop the transcript (or the audio + transcript pair, same basename) into imports — the Find phase picks it up. Nothing else is required.
  • Worker configured (integrations.transcription.enabled: true in bridge-config.yaml): Phase 0 pulls finished transcripts via the sync_script before scanning (pull), and the Find phase may hand un-transcribed audio back to the worker (push).
  • Fail-soft (hard rule): integration disabled → skip silently; worker unreachable (sync script exits non-zero) → proceed with whatever is already in imports. A missing or broken worker must never block a debrief run.

The full worker contract (config block, sync-script verbs, failure semantics, output format) lives in docs/transcription-worker.md.

Integration points

  • Transcription worker (optional, bring-your-own): see § Transcription pipeline above + docs/transcription-worker.md. A full reference implementation ships as skills/meeting-transcription/.
  • process-transcription (global, if installed): owns Org-specific participant lists and wiki routing — debrief defers to it.
  • project-advisor: governance + execution for GitHub issues created from extracted action items.
  • workflow/projects/{slug}.yaml: source of truth for issue field values.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.