backlog

A command for managing the rpm backlog, the persistent list of project tasks stored in an Org-mode text file. It can add, list, review, postpone, or complete tasks and archives finished entries.

In plain words
What is it for?
Use it to record new work, view tasks in execution order, inspect blocked or postponed items, and mark tasks complete.
Why use it?
It keeps long-term work separate from the temporary task list used in one agent session, so unfinished tasks are not lost.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/dppdppd/rpm/backlog
Any agent
npx skills add dppdppd/rpm --skill backlog
Clone the repo
git clone --depth 1 https://github.com/dppdppd/rpm

Made for: Claude Code, Codex.

Per session 156 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,882 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00156 $0.02882
Opus 5 $0.00078 $0.01441
Sonnet 5 $0.00031 $0.00576
Haiku 4.5 $0.00016 $0.00288

Measured 2d ago against content hash fae44255a37e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

backlog scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/backlog/SKILL.md · 256 lines

How it starts

The opening of the file, as written. The whole thing — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/backlog

Manage the rpm backlog — persisted at docs/rpm/future/tasks.org. All operations read and write this file using org-mode format.

Project Amendments

At the start of every invocation, check whether docs/rpm/skills/backlog.md exists in the consuming project. If it does, read it and apply its contents as additional project-specific instructions for this skill. Amendments may add band orderings, extra parent groups, or custom triage rules. They cannot remove or override plugin defaults — on conflict, this SKILL.md wins.

The backlog is sorted in execution order (top-to-bottom = the order in which tasks need to get done). "Take from the top" is the expected read pattern.

Within each * Parent group, keep this band order top-to-bottom:

  1. Actionable** IN-PROGRESS or ** TODO with all :BLOCKED_BY: deps DONE
  2. Blocked** BLOCKED, or ** TODO with unresolved :BLOCKED_BY:
  3. Postponed** TODO with a :POSTPONED: stamp

Closed entries (** DONE / ** CANCELLED) are archived to docs/rpm/future/done.org by /session-end and do not live in tasks.org long-term. /backlog done toggles TODO → DONE in place; the archive sweep runs at the next session-end.

Blocked and postponed items drift to the bottom of their band automatically whenever this file is written (during add, postpone, review, or session-end Phase 3b). Moves are mechanical — no user question. Preserve relative order within each band. New add entries land at the bottom of the actionable band (not the absolute bottom); user promotes upward explicitly if something needs to happen sooner.

Pivot capture (automatic, no user question). When the user redirects mid-session to new multi-step work that meets the TaskCreate bar, the LLM must insert a ** TODO at the top of the actionable band AND update docs/rpm/~rpm-session-start to set the task: field to the new work. The ask IS the confirmation — don't prompt. Skip for tactical single-step follow-ups. Rationale: the backlog should always represent current state, so a dropped session (without /session-end) still hands off accurately. Full rule lives in plugin/hooks/_directives.sh.

Read the full file on GitHub · 256 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 256 lines · 156 tokens per session scan A fae44255a37e

Subscribe to this mod's changes

backlog is a skill published in the GitHub repository dppdppd/rpm (3 stars, last pushed 4d ago), licensed MIT. It adds 156 tokens to every session and 2,882 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.