agentleFS
benororbenoror/obsidianos_work/.agents/skills/sprint-retro/SKILL.md

sprint-retro

Skill166 starsChanged 5 months agoMITView on GitHub

Draft a Sprint Retro by synthesizing scrum dailies, recaps, the sprint planning note, and the prior retro. Args: [sprint number | dates]. No args = infer the current sprint.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Always: use plain list items: - Description….

    line 29
  2. Very specific:

    Never leave an undated Sprint N Retro.md or a <Team> - Sprint Planning … variant.

    line 38
  3. Very specific:

    Prefer qmd over grep for discovery (see qmd).

    line 63
  4. Very specific:

    Leave Notes: empty for the user to paste the Confluence retro URL.

    line 126

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Sprint Retro

Synthesize a draft Sprint Retro from everything already in the vault — daily standups, weekly recaps, the sprint planning note, and the previous retro — so the live retro starts from a populated document instead of a blank template.

Usage

  • /sprint-retro — Infer the current/most-recent sprint from the prior retro + cadence.
  • /sprint-retro <N> — Draft the retro for sprint number N.
  • /sprint-retro <dates> — Use an explicit sprint window (e.g. 2026-05-19..2026-06-02).

Date Filtering

See date-filter for date-range syntax and parsing.

Output format (mandatory)

Mirror followup-todos — the retro is a bullet document, not a task list:

  • Never insert Obsidian Tasks / checkbox lines: no - [ ], - [x], - [-], - [?] — including in the Action items section.
  • Never append Tasks date tokens (📅, 🛫, ⏳). Write dates inline in prose / ISO (e.g. "by 2026-06-01").
  • A single trailing priority emoji (🔺 ⏫ 🔼 🔽) as plain text is fine when it reads naturally.
  • Always use plain list items: - Description…. Wikilinks ([[@Name]], [[Recap …]]), [PROJ-123](https://…) links, and bold are encouraged.

File naming convention (mandatory)

Sprint artifacts in Meetings/Engineering/Scrum/ follow a single pattern — keep it consistent:

  • Retro: Sprint <N> Retro <YYYY-MM-DD>.md (date = sprint end date / retro day).
  • Planning: Sprint <N> Planning <YYYY-MM-DD>.md (date = planning meeting day).

Never leave an undated Sprint N Retro.md or a <Team> - Sprint Planning … variant. If you encounter one while gathering context, flag it and offer to git mv it into the canonical form (updating any [[wikilinks]] that reference the old name).

Workflow

Step 1: Determine sprint number and window

  1. Verify the current date (see Date Verification in skill-conventions): date +%F.
  2. List Meetings/Engineering/Scrum/Sprint * Retro *.md to find the prior retro (Sprint <N-1> Retro <date>).
  3. Derive this sprint:
    • Number = prior + 1 (or the explicit <N> arg).
    • Window = the day after the prior retro's end through this sprint's retro day. Retros land on a biweekly cadence (same weekday, ~14 days apart) — extrapolate, then sanity-check against the planning note's date and the scrum dailies present.
  4. Confirm the resolved sprint number and start..end window with the user if anything is ambiguous (missing prior retro, irregular cadence, mid-sprint run).

Step 2: Gather sources

See vault-context. Gather in parallel:

SourceWhereWhat to extract
Scrum dailiesMeetings/Engineering/Scrum/YYYY-MM-DD.md within the windowYesterday/Today/Blockers, ticket IDs, merges, incidents, decisions
RecapsRecaps/Recap *.md whose period: overlaps the windowHighlights, decisions, open items, insights, cross-team notes
Sprint planningSprint <N> Planning <date>.mdCommitted tickets, estimates, decisions — to measure against outcomes
Prior retroSprint <N-1> Retro <date>.mdOpen Action items to carry forward and check off
TrackerTracker.mdCurrent ticket status (Done / Blocked / Paused / WIP)

Prefer qmd over grep for discovery (see qmd). Resolve names to [[@Name]] via people-resolver.

Step 3: Synthesize

Distill the gathered material into the retro sections, attributing each bullet to its source(s) with wikilinks (e.g. ([[Recap 2026-05-31]], [[Meetings/Engineering/Scrum/2026-05-28]])). Enrich bullets with Jira links ([PROJ-1234](https://your-org.atlassian.net/browse/PROJ-1234)) where a ticket is referenced.

  • Went Well / Start doing — shipped tickets, milestones, unblocks, demos.
  • Didn't Go Well / Stop doing — misses, recurring blockers, process friction, deadline pressure. Note cross-sprint patterns.
  • Keep doing / Shoutouts — one bullet per person, citing their concrete contributions. Only include people with evidence in the sources; don't invent contributions.
  • Discussion/Proposals — open questions, decisions to make, timelines.
  • Action items — concrete next steps (plain bullets, see output rules), including carried-over items from the prior retro.

Step 4: Compose the document

Base structure on Templates/Sprint Retro.md, with these additions used by recent retros:

---
Notes:
created: YYYY-MM-DDTHH:MM:SS±HH:MM
Participants: "[[+Engineering]]"
---

# Sprint <N> Retro — <YYYY-MM-DD>

**Sprint:** <N> · **<start>** → **<end>**

Planning context: [[Sprint <N> Planning <date>]]

> [!info] Draft synthesized from the vault
> Bullets below are distilled from **Scrum dailies** (`<start>`–`<end>`), the recaps covering this window, and the **[[Sprint <N> Planning <date>]]** commitments. Adjust after the live retro or paste a Confluence URL into `Notes:`.

### ▶️ Start doing / Went Well

- …

### 🛑 Stop doing / Didn't Go Well

- …

### 🙌 Keep doing / Shoutouts

- **[[@Name]]** — …

## Discussion/Proposals

- …

## Action items

- … 🔼
- Carry: … ⏫

## Sources (vault index)

| Kind | Notes |
| --- | --- |
| Recaps | [[Recap …]] (range), [[Recap …]] (range) |
| Scrum DS | [[Meetings/Engineering/Scrum/<start>]] … [[Meetings/Engineering/Scrum/<end>]] |
| Grooming / forums | [[Sprint <N> Planning <date>]] |
| Prior retro | [[Sprint <N-1> Retro <date>]] |

Leave Notes: empty for the user to paste the Confluence retro URL. Never set modified: manually (Obsidian manages it).

Step 5: Present and confirm

Show the draft to the user before saving. Let them adjust bullets, attributions, shoutouts, and action items.

Step 6: Save

Save to Meetings/Engineering/Scrum/Sprint <N> Retro <end-date>.md — see the naming convention above. If a retro for the sprint already exists, ask whether to overwrite.

Step 7: Offer to commit

See /commit.

Important Notes

  • Draft, not gospel — the output is a synthesis to seed the live retro; flag it as such in the [!info] callout and expect the user to revise.
  • Evidence-only shoutouts — only credit contributions that appear in the sources; omit team members with no traceable activity rather than guessing.
  • Cadence is a heuristic — biweekly same-weekday is the norm, but planning/retro days drift; always cross-check the derived window against the actual scrum dailies and planning-note date.
  • Naming consistency — if you touch sibling planning/retro files with non-canonical names, normalize them via git mv and fix referencing wikilinks (see the convention section).
  • No checkboxes — this is a discussion artifact; keep it bullets-only per the output rules, matching /followup-todos.
  • Timezone — use the vault owner's offset from USER.md. See skill-conventions Project Context.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.