agentleFS
bex-cobex-co/bex-security/.claude/skills/pm/SKILL.md

pm

Skill54 starsChanged 3 months agoApache-2.0View on GitHub

Inspect and maintain the repository's .pm workstreams, inbox notes, milestones, and tasks. Use when the user asks for PM board status or requests a supported board mutation such as creating, promoting, adding, or completing work.

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 0 are distinctive, and 10 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 10

    Very specific

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

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

    Respect the anti-goals: Read .pm/DO_NOT_DO.md before proposing or materializing work.

    line 24
  2. Very specific:

    Use worker: worker1 unless the workstream README names another worker.

    line 30
  3. Very specific:

    Milestones must be meaningful: Every milestone must include direct project-goal linkage, an observable expected outcome, and why this work …

    line 31
  4. Very specific:

    After editing any .md under .pm/, run npx prettier@3.4.2 --write ".pm//*.md" to keep the board consistently formatted.

    line 42
  5. Very specific:

    Read the tree (find .pm -type f -name '*.md', skipping done/) and .pm/DO_NOT_DO.md.

    line 48
  6. Very specific:

    Create the next free wN/ with a role-neutral README.md from the workstream template below.

    line 59
  7. Very specific:

    Create the next free inbox note wN/NNN.md with the idea as plain terse markdown (no frontmatter), with a one-line Why: … directly under the …

    line 63
  8. Very specific:

    If it is ≤ ~1h: do NOT create a milestone.

    line 70
  9. Very specific:

    Create the next tNNN.md from the task template and add its row to the milestone README.md table before the standing closing tasks, updating …

    line 74
  10. Very specific:

    If no open tasks remain in the milestone, move the whole milestone to wN/done/mN/ and check its box (- [x]) in the workstream README.md.

    line 81

≈ “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

Task: Arrange the .pm board

/pm is the only skill that writes to .pm/. It arranges milestones and tasks under the conventions below. /pm-brainstorm proposes; /pm materializes. This file is the canonical definition of the board conventions — hierarchy, sizing rule, quality gate, standing closing tasks, templates. /pm-brainstorm reads it at runtime and must not restate or diverge from it. Parse the subcommand from $ARGUMENTS (default = status).

The .pm hierarchy

LevelPathMeaningEffort
WorkstreamwN/ (w1, w2, …)a general-purpose worker queue; README.md + inbox notes—
Inbox notewN/NNN.md (w1/005.md)one idea or a sub-hour unit of work, plain markdown≤ ~1h
MilestonewN/mN/ (m1, m2, …)a shippable chunk: README.md + task files> ~1h, multiple tasks
TaskwN/mN/tNNN.mda single unittens of minutes

Rules (enforce every time)

  • Respect the anti-goals. Read .pm/DO_NOT_DO.md before proposing or materializing work. Do not create milestones/tasks that conflict with it.
  • Workers are general-purpose. Every wN may take work anywhere in the bex-security project. Workstream placement records scheduling and history, not a permanent mission, specialty, component boundary, or ownership claim. Do not assign themes or missions to workstreams, and do not use prior milestones to reject otherwise appropriate work.
  • Sizing rule. A milestone must be > ~1 hour of work across more than one task. If a chunk is ≤ ~1h (tens of minutes, a task or two), do NOT create an mN/ directory — record it as a loose inbox note wN/NNN.md. Tasks take tens of minutes; milestones take hours.
  • IDs must match the path. A task's id: wN/mN/tNNN frontmatter must equal the directory it lives in. Never create a milestone dir whose path disagrees with the IDs inside it — if you find such drift, flag/repair it, don't copy it.
  • Keep status in sync across all three places it lives: the workstream README.md milestone checkbox, the milestone README.md **Status:** line + the — DONE marker in the task table, and each task's status: frontmatter.
  • Numbering: next free zero-padded 3-digit for inbox notes (NNN) and tasks (tNNN); next free wN / mN. Scan the tree first; don't reuse a number.
  • Use worker: worker1 unless the workstream README names another worker.
  • Milestones must be meaningful. Every milestone must include direct project-goal linkage, an observable expected outcome, and why this work matters now (dependency/risk/sequence rationale).
  • Every board item briefly explains why. Milestones carry it in the milestone README.md (## Source + Goal linkage). Single tasks — inbox notes wN/NNN.md — carry it in the note itself: a one-line Why: … directly under the title (a sentence on the motivation/payoff is enough).
  • Every milestone ends with standing closing tasks, appended after the implementation tasks whenever a milestone is materialized:

    1. Surface parity — only when the milestone is feature development or a fix that touches the public CLI/SDK surface. Check that the change is consistent across every surface it exposes: CLI commands/flags/env vars and the TypeScript SDK in sdk/typescript/ (commands, arguments, flags, accepted values, and defaults are public API — see AGENTS.md "Public CLI changes"), the container (Dockerfile, docker/), and the docs (README.md, SECURITY.md, RELEASING.md). Update relevant help, schemas, documentation, and tests in the same change; flag any drift as follow-up work rather than silently diverging. Omit this task only for milestones with no public-surface change (pure infra, internal mechanism, docs, etc.) — note why it was omitted in the milestone's ## Source + Goal linkage.
    2. Simplify — run /simplify over the code this milestone changed (reuse / simplification / efficiency; behavior-preserving).
    3. Test coverage — add meaningful tests for the behavior this milestone shipped. Tests must assert real behavior and failure modes; never game coverage with trivial, tautological, or snapshot-everything tests (see sdk/typescript/TESTING.md).
    4. Closeout — the final task, added last. When the milestone's other tasks are all complete and its definition of done is actually met, close the milestone: set every remaining task's status: done, move each tNNN.md to wN/mN/done/, mark every row — **DONE** and set **Status:** done in the milestone README.md, move the whole wN/mN/ directory to wN/done/mN/, and check - [x] in the workstream README.md. Completing this task is the move — running /pm done <wN/mN/tNNN> on it last triggers the milestone move (the done subcommand's step 4). Do not run it until the DoD holds: a milestone lands in done/ when its observable end state is real, not merely when the code is written.

    Each depends_on the last implementation task(s) (Simplify and Test coverage depend on Surface parity when it's present; Closeout depends on Test coverage) and all count toward the (N tasks) total. add-task inserts new work before these (before Closeout) and updates their depends_on.

  • After editing any .md under .pm/, run npx prettier@3.4.2 --write ".pm/**/*.md" to keep the board consistently formatted.

Subcommands

status (default)

Read the tree (find .pm -type f -name '*.md', skipping done/) and .pm/DO_NOT_DO.md. Print, per open workstream: its milestones with **Status:**, and the next actionable task per milestone — the first non-done task whose depends_on are all satisfied. Also list open inbox notes. Then run a lightweight validation pass and flag:

  • items conflicting with .pm/DO_NOT_DO.md,
  • milestones missing ## Source + Goal linkage,
  • milestones whose definition of done is vague/non-testable,
  • inbox notes missing a Why: line.

Touch no files.

new workstream

Create the next free wN/ with a role-neutral README.md from the workstream template below.

add <wN> <idea…>

Create the next free inbox note wN/NNN.md with the idea as plain terse markdown (no frontmatter), with a one-line Why: … directly under the title. This is the default home for sub-hour work.

promote <wN/NNN> / new milestone <wN> <title>

Apply the sizing rule first.

  • If the work is > ~1h and splits into more than one task: create wN/mN/ with README.md (milestone template) + one tNNN.md per task (task template) + the standing closing tasks (Surface parity when it's feature dev/a fix touching the public CLI/SDK surface, then Simplify, then Test coverage, then Closeout), add the - [ ] **mN** — … line to the workstream README.md, and fill ## Source + Goal linkage with source + goal linkage + expected outcome + why-now rationale (note there why Surface parity was included or omitted).
  • If it is ≤ ~1h: do NOT create a milestone. Keep/append it as an inbox note wN/NNN.md (with its one-line Why: …) and tell the user why (too small for a milestone).

add-task <wN/mN> <title>

Create the next tNNN.md from the task template and add its row to the milestone README.md table before the standing closing tasks, updating their depends_on to include it. Update the (N tasks) count in the workstream README.

done <wN/mN/tNNN>

  1. Set the task's frontmatter status: done.
  2. In the milestone README.md: mark the row — **DONE** and update the **Status:** line (e.g. todo (t001 done)).
  3. Move the file to wN/mN/done/tNNN.md.
  4. If no open tasks remain in the milestone, move the whole milestone to wN/done/mN/ and check its box (- [x]) in the workstream README.md.

Show the intended moves before mutating if the user passed DRY_RUN=1.

Templates

Workstream README.md

# wN — project workstream (<worker>)

**Worker:** <worker>. This is a general-purpose bex-security workstream. It may take work anywhere in the project; the milestones below are scheduled work and historical records, not a permanent purpose, specialty, or ownership boundary.

## Milestones

- [ ] **mN** — <title> (<N> tasks) ← from <source>

Milestone README.md

# wN · mN — <name>

**Worker:** <worker> **Goal:** <what shipping this achieves> **Status:** todo

## Tasks (in order)

| id   | title   | est | depends_on |
| ---- | ------- | --- | ---------- |
| t001 | <title> | 30m | —          |

## Definition of done

<observable, testable end state>

## Source + Goal linkage

- **Source:** <pointer to the inbox note / brainstorm / docs this came from>
- **Goal linkage:** <which project goal / pillar this advances>
- **Expected outcome:** <observable impact after shipping>
- **Why now:** <dependency / risk / sequence rationale>

Task tNNN.md

---
id: wN/mN/tNNN
title: <title>
worker: <worker>
status: todo
estimate: 30m
depends_on: [wN/mN/tMMM]
---

## Objective

<one paragraph>

## Context

- <concrete paths / commands / facts>

## Steps

1. <step>

## Files

- <paths to touch>

## Acceptance criteria

- [ ] <testable check>

## Out of scope

- <deferred adjacent work>

Inbox note wN/NNN.md

Plain terse markdown, no frontmatter — one idea or a sub-hour unit of work. Opens with, directly under the title:

Why: <one sentence — motivation / payoff>

Arguments

$ARGUMENTS

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.