agentleFS
ByronScottJonesByronScottJones/Agents/skills/twg-status-rollups/SKILL.md

twg-status-rollups

Skill7 starsChanged 2 months agoMITView on GitHub

Use with root `twg` for status rollups, personal work summaries, and decision-readiness or go/no-go briefs. Routes to `pr-tree`, `org-tree`, `work-tree`, or `workitem-tree`.

The file

Show its rules and text

We split this skill into the rules it gives an agent

15 rules. About 0 are common, 9 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

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

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

    Use with the root twg skill.

    line 11
  2. Very specific:

    Run twg <command>. On shell command not found, use $HOME/.local/bin/twg (macOS/Linux) / $env:LOCALAPPDATA\Programs\twg\bin\twg.exe …

    line 16
  3. Very specific:

    Use pr-tree first. It groups by reporting-tree directReports.

    line 57
  4. Very specific:

    If option shape is uncertain, inspect twg help describe "pr-tree" before the data call.

    line 59
  5. Distinctive:

    Start count-first, then at most one supported sampling/full-fetch pass for repo or theme evidence, and synthesize at manager/team level.

    line 61
  6. Distinctive:

    Do not issue per-person queries to populate groups.

    line 63
  7. Distinctive:

    Add one secondary surface only for a named gap.

    line 65
  8. Distinctive:

    Start count-first on tree surfaces.

    line 72
  9. Very specific:

    Read Confluence page bodies with confluence content get, not docs get.

    line 77
  10. Distinctive:

    Stop when evidence is sufficient.

    line 79
  11. Distinctive:

    Do not list every artifact.

    line 143
  12. Distinctive:

    Do not infer goal/project health from issue counts alone.

    line 144
  13. Distinctive:

    Do not fan out per org member when manager/team grouping answers the prompt.

    line 145
  14. Distinctive:

    Do not use search snippets as final evidence for status or risk.

    line 146
  15. Very specific:

    Start explicit merged/open PR rollups with pr-tree, not the other trees.

    line 147

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

twg-status-rollups

Use with the root twg skill. Get exact command grammar from live twg help, twg help <terms>, or twg help describe <path>.

CLI launcher fallback

Run twg <command>. On shell command not found, use $HOME/.local/bin/twg (macOS/Linux) / $env:LOCALAPPDATA\Programs\twg\bin\twg.exe (PowerShell), then tell user to add that directory to PATH. Do not treat auth or command errors as PATH failures.

Use When

  • "What did I/person/team/org work on?", weekly update, standup, handoff, or restart after time away
  • "Status of project/goal/topic/focus area"
  • "Is this project ready to launch?" or go/no-go brief
  • Leadership, monthly, annual, cycle, appraisal, or goal-alignment readout
  • Org bottlenecks, priorities, stale goals, or project risks

First Move

Resolve scope before retrieval:

  • Person/team/org: resolve identity, roster, or org-tree groups.
  • Project, goal, focus area, or topic: resolve native key, ARI, URL, name, or a central page/workitem anchor.

Establish the time window. If absent, ask when precision matters; otherwise use a recent bounded window and state it. Keep personal summaries to a year or less.

Tree Routing Matrix

  • pr-tree: merged/open PRs, reviews, shipped work, or repo momentum by manager or team. Default for PR-constrained evidence.
  • org-tree: manager chain, roster, and reporting structure; grouping only, not delivery evidence.
  • workitem-tree: Jira issue load and active/done movement by org tree.
  • work-tree: multi-surface org counts across Jira, PRs, goals, projects, docs, videos. Not for PR-only or Jira-only asks.

Fast Path: PR-Based Leadership Rollup

For PR-based leadership prompts:

  1. Preserve the requested scope and window; for "last 2 weeks," use --since 14d.
  2. Use pr-tree first. It groups by reporting-tree directReports; add org-tree only for hierarchy context.
  3. If option shape is uncertain, inspect twg help describe "pr-tree" before the data call; do not probe incompatible flag combinations.
  4. Start count-first, then at most one supported sampling/full-fetch pass for repo or theme evidence, and synthesize at manager/team level. Do not issue per-person queries to populate groups; add one targeted PR follow-up only when a material theme lacks proof.
  5. Add one secondary surface only for a named gap.

Target 2-4 calls.

Evidence Policy

  • Match evidence to prompt constraints; merged-PR conclusions need PR evidence.
  • Start count-first on tree surfaces; hydrate examples only for themes, risks, or owner attribution.
  • Rank broad lists before minimal hydration.
  • Batch ranked reads with goals get, projects get, focus-areas get, jira workitem get, or pull-requests get.
  • Read Confluence page bodies with confluence content get, not docs get.
  • Distinguish authored delivery from review, coordination, and influence.
  • Stop when evidence is sufficient. After two identical backend failures, stop that path and report the gap.

Recipe Cards

Person Or Personal Update

Resolve the person, then pull recent Jira work, PRs, docs/pages, meetings, and project/goal involvement. Load references/personal-work-summary.md for subject, notification, PR hydration, and outcome-first rules. Separate delivery, review, docs/strategy, coordination, and influence.

Short-Window Personal Update / Standup

Load references/personal-work-summary.md. Resolve the person, preserve the requested project and window, prioritize material work, and separate evidence gaps from confirmed blockers.

Team Or Org Leadership Readout

Resolve org-tree first. Org projects: twg projects query --scope org --include-inferred ([Paid: Enriched]). Group results; hydrate outliers affecting momentum, blockers, or ownership.

Project Or Goal Status

Fetch the native project/goal first, with owner, state, update, links, dates, and recency. Hydrate only risk, progress, or dependency evidence.

Decision Readiness / Go-No-Go

Load references/decision-readiness.md. Resolve the native project or decision anchor first and keep explicit links as the scope boundary. Identify the gates, then give the decision or recommendation with confidence, gaps, and change conditions. Do not infer owners.

Topic Status

Resolve/search once, select central project, goal, page, or workitem anchors, then hydrate those before broad work/activity queries.

References

  • references/personal-work-summary.md - standups, catch-ups, and broader personal status evidence
  • references/decision-readiness.md - cross-domain go/no-go and approval evidence

Appraisal / Performance Evidence

Resolve person and horizon. Separate delivery, review, collaboration, docs/strategy, project/goal impact, and stakeholder signals. Avoid count-only ranking; caveat weak evidence.

Output Shape

  • Executive summary first, with 3-6 high-signal observations.
  • Table with owner/team/workstream, positive and risk signals, current focus, confidence, and evidence.
  • Risks ranked by impact and owner.
  • Confidence and gaps: stale updates, missing coverage, ACL gaps, or sampling boundaries.

Anti-Patterns

  • Do not list every artifact.
  • Do not infer goal/project health from issue counts alone.
  • Do not fan out per org member when manager/team grouping answers the prompt.
  • Do not use search snippets as final evidence for status or risk.
  • Start explicit merged/open PR rollups with pr-tree, not the other trees.

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.