** (`<id>`, P<n>) — <status; latest note if useful> ## Detail  ### <Initiative title> (P<n>) <n epics, n tasks>. - **<epic title>** — <distilled close_reason> - **<epic title>** — <distilled close_reason> --- *Source: <n> flush events between <date> and <date>. <If excluded:> <n> throwaway issue(s) excluded.* ``` ### Mode rules - **Detailed mode**: include the "Detail" section with per-initiative breakdown. - **Summary mode**: skip "Detail" entirely. Just Highlights + Currently in flight. - **Currently in flight**: omit the section if no open/in-progress issues. Show priorities and IDs in detailed mode; collapse to a one-line count in summary mode if there are many (>5). - **Empty period**: say so cheerfully — "Quiet week — nothing flushed, nothing in flight." ### Tone - Friendly, conversational. Aide-mémoire vibe — "oh yeah, *that's* what I was doing." - Distill close_reasons; don't quote them verbatim. - No emoji. - Test counts **only** if they signal a problem ("3 failing tests", "still need to write tests for X"). Don't trumpet "457 tests passing". - Dates rough ("all on 2026-04-16", "shipped Wednesday") — not minute-precise. - If a flush event has its own `summary` field, that's the user's curated editorial — lean on it for the headline. --- ## Multi-project mode Find candidate databases with `Glob`: ``` <scan-dir>/**/.ait/ait.db ``` For each, run the queries with `--db <path>`: ```bash ait --db <path> config ait --db <path> log --since YYYY-MM-DD --long ait --db <path> list --all --long ``` Skip projects with no activity in the period (no flushed items, no live issues touched in range). ### Output template ```markdown # Recent Activity Across Projects (<start-date> → <today>) <n> projects with activity in the period. ## Overall highlights - **<prefix>**: <one-line of what shipped> - **<prefix>**: <one-line of what shipped> ## Per-project ### `<prefix>` — `<path>` <single-project body, without its own top-level heading> ### `<prefix>` — `<path>` <...> ``` ### Mode rules - **Detailed mode**: include "Per-project" with each project's full single-project body (Highlights + Currently in flight + Detail). - **Summary mode**: drop "Per-project" entirely — overall highlights only. - **Empty scan**: "No activity across <n> projects in the period — must've been on holiday." --- ## Tips - The user's `ait` agent name in this repo (and probably others) may already be set in memory — check before claiming any issues, but this skill is **read-only** so you shouldn't need to claim anything. - If `ait config` returns no prefix, the project hasn't been initialised — flag it and move on. - For multi-project mode against ~10+ projects, the raw JSON could grow context. If it gets unwieldy, summarise each project's data into the markdown line by line as you go, rather than holding it all at once. --- > Source: [ohnotnow/agent-issue-tracker](https://github.com/ohnotnow/agent-issue-tracker) — distributed by [TomeVault](https://tomevault.io). <!-- tomevault:4.0:skill

# Ohnotnow Agent Issue Tracker Ait Recap > AIT Recap - Skill: `tomevault-io/ohnotnow-agent-issue-tracker-ait-recap` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/ohnotnow-agent-issue-tracker-ait-recap` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/ohnotnow-agent-issue-tracker-ait-recap/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/ohnotnow-agent-issue-tracker-ait-recap --- # AIT Recap Build a friendly, human-readable markdown summary of recent activity from the `ait` issue tracker. The report **is** the deliverable — present the markdown directly to the user. ## Resolve the request Three things to pin down from the user's prompt: ### 1. Period (default: this week) - "today" / "since this morning" → today's date - "this week" / "last week" / no period given → 7 days ago - "this month" / "last month" → 30 days ago - Explicit date ("since 2026-04-01", "last fortnight") → use as-is Convert to a `YYYY-MM-DD` for `ait log --since` and client-side timestamp filtering. ### 2. Mode (default: detailed) - Prompt contains "summary", "summarise", "tl;dr", "high level", "quick", "brief" → **summary mode** - Otherwise → **detailed mode** ### 3. Scope (default: current project) - Prompt mentions a directory ("scan `~/Documents/code/`", "all my projects in X", "across everything") → **multi-project mode** - Otherwise → **single project mode** --- ## Single-project mode Run these queries: ```bash ait config # project prefix ait log --since YYYY-MM-DD --long # flushed history in the period ait list --all --long # all live issues (filter by date client-side) ``` For live issues, keep any with `created_at`, `updated_at` or `closed_at` >= the period start. ### Throwaway filter Drop flushed items where **both**: - `close_reason` is null or empty, **and** - the item has no children (no other items in the same flush list have it as `parent_id`). This silently culls test detritus while keeping deliberate-but-unnoted closes visible. ### Output template ```markdown # This in `` ( → ) ## Highlights - **** — - **<...>** - **<...>** ## Currently in flight - **** (`<id>`, P<n>) — <status; latest note if useful> ## Detail  ### <Initiative title> (P<n>) <n epics, n tasks>. - **<epic title>** — <distilled close_reason> - **<epic title>** — <distilled close_reason> --- *Source: <n> flush events between <date> and <date>. <If excluded:> <n> throwaway issue(s) excluded.* ``` ### Mode rules - **Detailed mode**: include the "Detail" section with per-initiative breakdown. - **Summary mode**: skip "Detail" entirely. Just Highlights + Currently in flight. - **Currently in flight**: omit the section if no open/in-progress issues. Show priorities and IDs in detailed mode; collapse to a one-line count in summary mode if there are many (>5). - **Empty period**: say so cheerfully — "Quiet week — nothing flushed, nothing in flight." ### Tone - Friendly, conversational. Aide-mémoire vibe — "oh yeah, *that's* what I was doing." - Distill close_reasons; don't quote them verbatim. - No emoji. - Test counts **only** if they signal a problem ("3 failing tests", "still need to write tests for X"). Don't trumpet "457 tests passing". - Dates rough ("all on 2026-04-16", "shipped Wednesday") — not minute-precise. - If a flush event has its own `summary` field, that's the user's curated editorial — lean on it for the headline. --- ## Multi-project mode Find candidate databases with `Glob`: ``` <scan-dir>/**/.ait/ait.db ``` For each, run the queries with `--db <path>`: ```bash ait --db <path> config ait --db <path> log --since YYYY-MM-DD --long ait --db <path> list --all --long ``` Skip projects with no activity in the period (no flushed items, no live issues touched in range). ### Output template ```markdown # Recent Activity Across Projects (<start-date> → <today>) <n> projects with activity in the period. ## Overall highlights - **<prefix>**: <one-line of what shipped> - **<prefix>**: <one-line of what shipped> ## Per-project ### `<prefix>` — `<path>` <single-project body, without its own top-level heading> ### `<prefix>` — `<path>` <...> ``` ### Mode rules - **Detailed mode**: include "Per-project" with each project's full single-project body (Highlights + Currently in flight + Detail). - **Summary mode**: drop "Per-project" entirely — overall highlights only. - **Empty scan**: "No activity across <n> projects in the period — must've been on holiday." --- ## Tips - The user's `ait` agent name in this repo (and probably others) may already be set in memory — check before claiming any issues, but this skill is **read-only** so you shouldn't need to claim anything. - If `ait config` returns no prefix, the project hasn't been initialised — flag it and move on. - For multi-project mode against ~10+ projects, the raw JSON could grow context. If it gets unwieldy, summarise each project's data into the markdown line by line as you go, rather than holding it all at once. --- > Source: [ohnotnow/agent-issue-tracker](https://github.com/ohnotnow/agent-issue-tracker) — distributed by [TomeVault](https://tomevault.io).