agentleFS
hushh-labshushh-labs/hushh-research/.agents/skills/planning-board/SKILL.md

planning-board

Skill40 starsChanged 4 days agoApache-2.0View on GitHub

Use when inspecting, summarizing, creating, triaging, or updating work on the Hussh Engineering Core (Project 73) or Hussh Action Items (Project 79) GitHub boards.

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. About 0 are common, 4 are distinctive, and 9 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 9

    Very specific

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

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

    Avoid overlap with repo-operations and repo-context.

    line 13
  2. Very specific:

    Verify the active writer identity before mutation.

    line 54
  3. Very specific:

    When creating new work, create the issue before its board item.

    line 56
  4. Very specific:

    When reporting work back to the user, always use the stable task shape #<number> <title>.

    line 59
  5. Distinctive:

    When the user asks for labels, treat labels as a first-class part of the task update instead of leaving them implicit.

    line 60
  6. Distinctive:

    Do not move sprint or reset dates on existing tasks unless the user explicitly asks for that metadata change.

    line 61
  7. Distinctive:

    Do not use bare issue numbers in summaries, status lists, overdue lists, or change logs when the title is available.

    line 62
  8. Very specific:

    When implementation is complete but acceptance remains, use the selected board's review state: In review on Project 73 or Ready for QA on …

    line 63
  9. Very specific:

    Use Done only when the work is accepted and the issue state or acceptance evidence agrees with completion.

    line 64
  10. Very specific:

    Before and after broad cleanups, run board_ops.py audit-state and resolve drift deliberately.

    line 65
  11. Very specific:

    If the task is broad repo orientation, start with repo-context.

    line 70
  12. Very specific:

    If the task is CI, deploy, or branch protection, use repo-operations.

    line 71
  13. Distinctive:

    If the task becomes product coding or docs governance, route to the correct owner skill for that domain.

    line 72

≈ “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

Hussh Planning Board Skill

Purpose and Trigger

  • Primary scope: planning-board-intake
  • Trigger on Hussh Engineering Core or Hussh Action Items board summaries, issue-backed board item creation, triage, and field updates.
  • Board selection: board_ops.py --board engineering-core (default, Project 73) or --board action-items (Project 79, the Inbox → Accepted → In Progress → Ready for QA → Done lifecycle). Resolve statuses and fields against the live catalog; see the per-board references.
  • Avoid overlap with repo-operations and repo-context.

Coverage and Ownership

  • Role: owner
  • Owner family: planning-board

Owned repo surfaces:

  1. .agents/skills/planning-board

Non-owned surfaces:

  1. repo-operations
  2. repo-context

Do Use

  1. Summarizing board activity by date window or repository slice.
  2. Creating issue-backed board work in hushh-labs/hushh-research.
  3. Updating required board metadata and verifying the resulting project-item state.
  4. Normalizing or updating issue labels when the user asks for label hygiene.
  5. Writing implementation-task handoffs for human owners and coding agents.

Do Not Use

  1. CI, deploy, branch-protection, or PR-approval workflows.
  2. Broad repo-orientation work.
  3. Product implementation or docs-home governance.

Read First

  1. .agents/skills/planning-board/references/engineering-core-board.md
  2. .agents/skills/planning-board/references/action-items-board.md
  3. docs/development/repository-map.md
  4. docs/operations/coding-agent-mcp.md
  5. .agents/skills/planning-board/references/task-handoff.md

Workflow

  1. Resolve live issue/project state and existing work before creating another task.
  2. Verify the active writer identity before mutation; honor the user's CLI/connector preference. A CLI requirement means using the authenticated gh identity, not connector credentials.
  3. For implementation tasks, follow references/task-handoff.md: link verified owner/lead profiles; require a reasoned human brief with the owner's review reminder and actual review status; use For coding agents for implementation mechanics and acceptance evidence.
  4. When creating new work, create the issue before its board item; otherwise edit the canonical issue in place. For Action Items, intake starts in Inbox; use the selected board reference for acceptance, ownership, and scheduling rules.
  5. Re-read the issue and board state after editing; verify people/reference links, preserved decisions and actual field values rather than trusting cached output.
  6. When reporting work back to the user, always use the stable task shape #<number> <title>.
  7. When the user asks for labels, treat labels as a first-class part of the task update instead of leaving them implicit.
  8. Do not move sprint or reset dates on existing tasks unless the user explicitly asks for that metadata change.
  9. Do not use bare issue numbers in summaries, status lists, overdue lists, or change logs when the title is available.
  10. When implementation is complete but acceptance remains, use the selected board's review state: In review on Project 73 or Ready for QA on Project 79. Keep unfinished implementation active; never invent a status option or infer deployment acceptance from source completion.
  11. Use Done only when the work is accepted and the issue state or acceptance evidence agrees with completion.
  12. Before and after broad cleanups, run board_ops.py audit-state and resolve drift deliberately.
  13. For duplicate board tasks, preserve scope and traceability in the canonical issue before removing the duplicate project item with board_ops.py remove-task. Use the description when comments are disallowed; do not mark duplicates as Done.

Handoff Rules

  1. If the task is broad repo orientation, start with repo-context.
  2. If the task is CI, deploy, or branch protection, use repo-operations.
  3. If the task becomes product coding or docs governance, route to the correct owner skill for that domain.

Required Checks

python3 -m py_compile .agents/skills/planning-board/scripts/board_ops.py
python3 .agents/skills/planning-board/scripts/board_ops.py summary --from 2026-04-01 --to 2026-04-07
python3 .agents/skills/planning-board/scripts/board_ops.py audit-state

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.