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.
npx agentmods add skills/ij5a/subrosa/checkpoint-backlognpx skills add ij5a/subrosa --skill checkpoint-backloggit clone --depth 1 https://github.com/ij5a/subrosaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ij5a/subrosa/checkpoint-backlog)<a href="https://agentmods.dev/skills/ij5a/subrosa/checkpoint-backlog"><img src="https://agentmods.dev/badge/skills/ij5a/subrosa/checkpoint-backlog.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00074 | $0.01919 |
| Opus 5 | $0.00037 | $0.00959 |
| Sonnet 5 | $0.00015 | $0.00384 |
| Haiku 4.5 | $0.00007 | $0.00192 |
Grade A, and why
checkpoint-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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
checkpoint-backlog: checkpoint queued sessions
When a session ends, subrosa's SessionEnd hook adds it to pending-checkpoint.log in ~/.claude/subrosa/ by default. This skill processes the queue in-session. It uses no background daemon or headless claude run. It applies the checkpoint skill to each past session.
Follow the checkpoint skill's 4 types, user, feedback, project, and reference rules. Follow its exclusion list and leaf to subrosa fact upsert to subrosa generate flow. Read ${CLAUDE_PLUGIN_ROOT}/skills/checkpoint/SKILL.md for details. Apply these overrides.
When the queue spans more than one project, read each session dump in parallel, with one sub-agent per project. Keep a single-project queue sequential. Separate projects have separate MEMORY.md files and do not race. Sessions in the same project would race regeneration and deduplication, so keep them serial.
Procedure
-
List the backlog: Run
subrosa pending. Each line is<timestamp>\t<session-id>, oldest first. Collect unique ids. If it is empty, say "no backlog" and stop. -
Cap the work at the 5 most recent queued sessions. The newest is last in the file. If more remain, process those 5. Tell the user to run
/subrosa:checkpoint-backlogagain for the rest. This prevents session startup from stalling. -
Find each session's project. For each id, run
subrosa session <id> | head -2. The pipe stops after 2 header lines. It avoids dumping the whole session and works with any recent binary:- Line 1 is
# session <id> project=<project> cwd=<cwd> <first>..<last>. Take theproject=value. - Line 2 is
# memdir: <path>. Take the memdir path.
If it prints
[subrosa] no archived turns for session <id>, the session was never ingested. There is nothing to extract. Runsubrosa checkpoint-drop <id>. Count it as a skip. Do not pass it to a lane. - Line 1 is
-
Group surviving ids by project, then choose a branch. If none survive, skip to the report.
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.
- 4d ago First seen · 104 lines · 74 tokens per session scan A fa7ac0bed3c8
checkpoint-backlog is a skill published in the GitHub repository ij5a/subrosa (7 stars, last pushed 4d ago), licensed MIT. It adds 74 tokens to every session and 1,919 once invoked, about $0.0004 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.
Other skills, from other repositories
memory-recall
Search and recall relevant memories from past sessions via memsearch. Use when the user's question could benefit from historical context, past decisions, debugging notes, previous conversations, or project knowledge -- especially questions like 'what did I decide about X', 'why did we do Y', or 'have I seen this…
marshal
Meta-orchestrator that takes any direction — broad, specific, or vague — and autonomously chains skills and context into actionable work. Gathers context from codebase, docs, and memory. Only asks the user when it genuinely cannot proceed. Single-session orchestrator.
wiki
Markdown-first knowledge base where the LLM acts as librarian. Ingests raw sources, compiles and interlinks topic files, self-maintains an index. No vector DB or embeddings required -- uses LLM-native navigation over structured markdown up to 400K words.
alive:session-history
Revive sessions (quick or heavy), browse, and search — 'what happened recently?', 'find the session where we discussed X', 'revive yesterday's session'. For single-session recall and multi-session browsing. If the human needs to merge multiple sessions into one working context or detect conflicts between parallel…
unslop-file
Humanize natural-language memory files (CLAUDE.md, todos, preferences, docs) by removing AI-isms and adding burstiness while preserving every code block, URL, path, command, and heading exactly. Two modes: --deterministic (fast, regex-based, no API) and LLM (default, calls Claude for rewrite). Humanized version…
alive-inbox
Scan 03Inbox/ for unrouted files, present routing suggestions.