agentleFS
egdev6egdev6/discord-project-manager/skills/x-queue-planner/SKILL.md

x-queue-planner

Skill4 starsChanged 5 months agoMITView on GitHub

Define a reviewable X queue planning slice from approved context or source ingestion before any persistence.

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. About 0 are common, 14 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈14

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill to turn approved project context or source ingestion into a small X queue plan that humans can review before any drafting, …

    line 9
  2. Distinctive:

    Start from approved brand context and recent ledger context when they are available.

    line 39
  3. Distinctive:

    Treat source ingestion as a normalization step that produces proposal-only queue candidates, never a durable write.

    line 40
  4. Distinctive:

    Keep X queue planning separate from cross-network strategy rules.

    line 41
  5. Very specific:

    If analytics or trends are supplied, use only sanitized normalized snapshots and trend signals from the shared …

    line 43
  6. Distinctive:

    Keep each queued item small enough for human review before any drafting or publishing.

    line 44
  7. Distinctive:

    Treat memory writes as planned targets until a human approves the queue.

    line 45
  8. Distinctive:

    Use only fake/demo values in repository-facing examples.

    line 47
  9. Distinctive:

    Follow ADR 0002 for all namespace references.

    line 48
  10. Very specific:

    Do not write or revise durable X queue planning memory until a human approves the queue.

    line 151
  11. Distinctive:

    Do not claim memory was written unless the runtime actually saved it after approval.

    line 171
  12. Distinctive:

    Do not consume raw scraped pages, screenshots, native exports, provider payloads, account data, credentials, or private metrics.

    line 172
  13. Distinctive:

    Do not claim live browser/API scraping, analytics ingestion, or trend-provider access until a separate approved implementation exists.

    line 173
  14. Distinctive:

    Do not publish or schedule content from this contract alone.

    line 174
  15. Distinctive:

    Do not put durable X queue planning under runtime Discord namespaces.

    line 175
  16. Distinctive:

    Do not include private brand plans, secrets, or real customer data in repo examples.

    line 176

≈ “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

X Queue Planner

Use this skill to turn approved project context or source ingestion into a small X queue plan that humans can review before any drafting, publishing, scheduling, or durable memory write happens.

This skill defines a planning contract only. It does not publish, schedule, sync to Buffer, route Discord traffic, or claim that memory writes are already wired.

Inputs

Required inputs:

  • project_slug
  • network_slug: x
  • timeframe
  • topics
  • cadence
  • tone
  • constraints
  • known_assets or none

Optional inputs:

  • brand_context_summary
  • recent_ledger_summary
  • normalized_snapshots for network: x
  • trend_signals scoped to X or cross-network strategy
  • queue_goal
  • review_preferences
  • source_context
  • source_ingestion

Behavior

  1. Start from approved brand context and recent ledger context when they are available.
  2. Treat source ingestion as a normalization step that produces proposal-only queue candidates, never a durable write.
  3. Keep X queue planning separate from cross-network strategy rules.
  4. Separate confirmed facts, assumptions, variation rules, and proposed angles or candidate updates.
  5. If analytics or trends are supplied, use only sanitized normalized snapshots and trend signals from the shared docs/operations/social-strategy-analytics-trends-ingestion.md contract; join X metrics through content_ledger_entry_id, keep unknown metrics as unknown, and include trend provenance, observed timestamp, confidence, and source type in the planning basis.
  6. Keep each queued item small enough for human review before any drafting or publishing.
  7. Treat memory writes as planned targets until a human approves the queue.
  8. For write-like Discord requests, load skills/discord-approval-gate/SKILL.md and ask for explicit approval before any persistence.
  9. Use only fake/demo values in repository-facing examples.
  10. Follow ADR 0002 for all namespace references.

Output shape

Return a YAML-like structure similar to this:

project: <project-slug>
network: x
timeframe: <timeframe>
source_context:
  brand_namespace_key: <brand namespace>
  strategy_namespace_key: <strategy namespace>
  ledger_namespace_key: <content ledger namespace>
  network_namespace_key: <x namespace>
source_ingestion:
  source_type: <fake-manual-input|approved normalized source>
  normalized_signals:
    - <reviewable source cue>
planning_inputs:
  topics:
    - <topic>
  cadence:
    items_per_week: <count>
    preferred_publish_days:
      - <day>
  tone:
    - <tone cue>
  constraints:
    - <constraint>
  known_assets:
    - <asset>
planning_basis:
  confirmed_facts:
    - <approved fact from source context>
  assumptions:
    - <planning assumption to review>
  variation_rules:
    - <queue diversity rule>
  sanitized_performance_inputs:
    - content_ledger_entry_id: <x ledger id or none>
      insight: <normalized analytics cue>
  trend_inputs:
    - source_type: <rss|web|search|trend-provider|manual-sanitized-entry>
      observed_at: <timestamp>
      confidence: <0.0-1.0>
      provenance:
        source_kind: <rss-feed|webpage|search-result|trend-report|manual-note>
        retrieved_by: <fake adapter or human-sanitized-entry>
        retrieval_mode: fake-fixture-no-network
        canonical_url: https://example.invalid/<path>
        source_title: <bounded title>
      strategy_signal: <normalized trend cue>
  proposed_angles:
    - <queued angle under consideration>
queue_plan:
  status: <draft-review|approved-for-demo-validation>
  entries:
    - id: <stable-demo-id>
      entry_type: <single-post|thread>
      working_title: <title>
      objective: <why this item exists>
      rationale:
        ties_to_topic: <topic linkage>
        brand_alignment: <approved context>
        ledger_reference: <content history cue or none>
      variation_note: <how this item differs from adjacent queue items>
      required_assets:
        - <asset>
      approval_checkpoint:
        - <human review step>
approval:
  status: <pending-human-approval|approved-for-demo-validation>
  checkpoints:
    - <approval rule>
approval_request:
  state: <approval-requested>
  reply_options:
    - approve write
    - "revise: <instruction>"
    - reject
memory_write_targets:
  project_strategy_namespace_key: <strategy namespace>
  network_namespace_key: <x namespace>
  write_mode: <planned-only-until-approved>

Memory behavior

Read candidates

  • discord-project-manager/project/<project-slug>/brand
  • discord-project-manager/project/<project-slug>/strategy
  • discord-project-manager/project/<project-slug>/content-ledger
  • discord-project-manager/project/<project-slug>/network/x

Write candidates

  • reusable cross-network queue-planning rules under discord-project-manager/project/<project-slug>/strategy
  • approved X-local queue planning context under discord-project-manager/project/<project-slug>/network/x

Approval gate

Do not write or revise durable X queue planning memory until a human approves the queue. Source ingestion must stop at proposal-only queue candidates until the operator replies with approve write.

Namespace target

Use ADR 0002 exactly:

  • discord-project-manager/project/<project-slug>/strategy
  • discord-project-manager/project/<project-slug>/network/x

Canonical ADR examples that this skill may mirror when using fake/demo values:

  • discord-project-manager/project/egdev/strategy
  • discord-project-manager/project/egdev/network/x

Promotion to repo artifact

Promote reusable X queue planning rules, approval checkpoints, and contract changes into repo artifacts when they become canonical, review-facing, or architecture-relevant.

Safety rules

  • Do not claim memory was written unless the runtime actually saved it after approval.
  • Do not consume raw scraped pages, screenshots, native exports, provider payloads, account data, credentials, or private metrics; require sanitized normalized snapshots instead.
  • Do not claim live browser/API scraping, analytics ingestion, or trend-provider access until a separate approved implementation exists.
  • Do not publish or schedule content from this contract alone.
  • Do not put durable X queue planning under runtime Discord namespaces.
  • Do not include private brand plans, secrets, or real customer data in repo examples.

Demo example (fake)

project: egdev
network: x
timeframe: 2026-W24
source_context:
  brand_namespace_key: discord-project-manager/project/egdev/brand
  strategy_namespace_key: discord-project-manager/project/egdev/strategy
  ledger_namespace_key: discord-project-manager/project/egdev/content-ledger
  network_namespace_key: discord-project-manager/project/egdev/network/x
planning_inputs:
  topics:
    - technical specificity in short-form posts
  cadence:
    items_per_week: 3
    preferred_publish_days:
      - monday
      - wednesday
      - friday
  tone:
    - direct
    - technical
  constraints:
    - no publishing without human approval
  known_assets:
    - asset://demo/x-thread-outline-002
planning_basis:
  confirmed_facts:
    - approved brand context prefers practical technical framing
  assumptions:
    - short threads will outperform generic single-line hooks
  variation_rules:
    - alternate between single-post and thread entries
  proposed_angles:
    - one queue item about reviewability discipline
queue_plan:
  status: draft-review
  entries:
    - id: x-queue-item-01-demo
      entry_type: thread
      working_title: reviewability thread demo
      objective: open a technical discussion without hype framing
      rationale:
        ties_to_topic: reinforces technical specificity
        brand_alignment: explain trade-offs before implementation
        ledger_reference: x-post-001-demo follow-up angle
      variation_note: longer thread to avoid repeating a prior single-post hook
      required_assets:
        - asset://demo/x-thread-outline-002
      approval_checkpoint:
        - approve the queue entry before drafting
approval:
  status: pending-human-approval
  checkpoints:
    - approve queue entries before drafting or publishing
memory_write_targets:
  project_strategy_namespace_key: discord-project-manager/project/egdev/strategy
  network_namespace_key: discord-project-manager/project/egdev/network/x
  write_mode: planned-only-until-approved

This example is fake/demo data only and must not be treated as a live editorial plan.

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.