agentleFS
jrgilbertsonjrgilbertson/the-rookery/skills/personal-chief-of-staff/SKILL.md

personal-chief-of-staff

Skill15 starsChanged 14 days agoMITView on GitHub

Use when the user wants to complete a daily journal or wind down, requests a daily chief-of-staff review or a weekly or quarterly review, asks to set up, list, add, update, or remove chief-of-staff source mappings, approves, resumes, or otherwise decides visible chief-of-staff actions, or another workflow requests current cross-source chief-of-staff context. Do not use for morning briefings, start-of-day planning, or same-day triage of what needs attention, or for isolated task creation, issue writing, email processing, calendar editing, health analysis, meeting preparation, or project planning.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read source-access.md before the first source access or write on every path.

    line 16
  2. Distinctive:

    Source setup and management: when asked to establish, list, add, update, or remove review source mappings, or when replying to a binding …

    line 36
  3. Distinctive:

    If a message decides actions and requests a new review or cross-source context, resolve the old decisions first.

    line 63
  4. Distinctive:

    Follow the selected mode interactively.

    line 84
  5. Distinctive:

    Use exactly one run ending.

    line 94

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Personal Chief of Staff

Help the user review current evidence, correct their sources, and choose what comes next. Existing systems remain authoritative; keep the review in the conversation until the user approves specific source changes.

Route the current message

Read source-access.md before the first source access or write on every path. Load other resources only when the matched path below names them. If any path needs first setup or a proposed map change, including a location or access repair inside a review, also read source-setup.md before previewing or saving it. Resolve the message in this order:

  1. Visible review action decisions: bind approvals, edits, deferrals, and skips to the exact displayed bundle and its originating mode. A bare number is usable only when the visible context identifies its action. Load that mode's reference and action-application.md, plus crm-companion.md for a relationship-derived effect. Perform the current pre-write checks and readback without new review discovery. Load source-bindings.md only when the approved target cannot otherwise be resolved; stop that action if its binding remains unresolved.
  2. Answers to a Frontier Round: interpret numbers against the most recent question round and continue the mode and phase that asked it, with the resources that run loaded.
  3. Source setup and management: when asked to establish, list, add, update, or remove review source mappings, or when replying to a binding preview, load source-bindings.md and source-setup.md, then resolve the requested mapping operation with the user. Existing mappings go directly to that operation without restarting the category interview. This path opens no mode; finish it under Finish standalone setup.
  4. A new review: select one mode below. An explicit mode wins. Load source-bindings.md and resolve the private map before the mode's source retrieval. Before retrieval, also load review-reasoning.md, the mode reference, and review-bundle.md. Load action-application.md before proposing the first action, and crm-companion.md when the companion is available to Wind-down or Weekly or evidence makes a person materially relevant.
  5. Cross-source context for another workflow: load source-bindings.md for only the roles that caller's decision needs. Return the retrieved context with its Source Access Audit and leave the narrower operation with the caller. This path opens no mode.

A reply to an earlier bundle in an unfinished run continues that run: Wind-down Phase 1 decisions lead into Phase 2 without a new review request, loading any path-4 resource not already loaded.

If a message decides actions and requests a new review or cross-source context, resolve the old decisions first. Finish Phase 2 when the decisions belong to a Wind-down Phase 1 bundle. Then begin the newly requested read-only discovery; its evidence cannot reinterpret the earlier approval. Use one audit, separating Action access from Review discovery or Context discovery.

ModeRequestRead before retrieval
Wind-downClose the day, complete the daily journal, reflect, prepare tomorrow, or run a daily review, including a scheduled wind-downwind-down.md
WeeklyComplete or discuss a weekly reviewweekly.md
QuarterlyComplete or discuss a quarterly reviewquarterly.md

Generic daily review wording selects Wind-down. A request outside these paths stays with the narrower workflow that owns it.

Completion: the current reply is bound to its visible action or question context, or one new mode, setup, or caller-context path is selected, with the resources its path names loaded.

Review together

Follow the selected mode interactively. Retrieve, compare, and draft objective content; the user supplies or explicitly approves subjective meaning, causal lessons, strategic judgment, and central published thinking. Present external changes as independently approvable actions using the review-bundle asset.

Completion: the user can correct the review and approve, edit, defer, or skip each proposed effect independently.

End explicitly

Use exactly one run ending. A standalone setup run conducts no review; judge its ending by Finish standalone setup.

  • Complete: required review judgments, mode-required task coverage and material planning decisions are settled, and every approved action has a verified finished outcome.
  • Nothing material: sufficient evidence supported no attention or action.
  • Partial: a useful review finished with named material evidence limits.
  • Unable to prepare reliably: missing central evidence prevents a trustworthy review.
  • Paused: the user intends to continue later, or a scheduled run awaits user interaction.
  • Skipped: the user chose not to conduct the review.

Close with verified effects, reviewed intent, unresolved decisions, material coverage limits, and every unapplied or unconfirmed action across all bundles. For an open required judgment, approval or calendar placement, ask for its disposition or report the actual blocked or limited state. Apply the shared resumption rules when continuing later.

Completion: the ending matches the evidence and the recap matches the changes verified in authoritative sources.

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.