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 skills add kalinbogatzevski/captain-memo --skill statsgit clone --depth 1 https://github.com/kalinbogatzevski/captain-memoWrote 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/kalinbogatzevski/captain-memo/stats)<a href="https://agentmods.dev/skills/kalinbogatzevski/captain-memo/stats"><img src="https://agentmods.dev/badge/skills/kalinbogatzevski/captain-memo/stats/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kalinbogatzevski/captain-memo/stats"><img src="https://agentmods.dev/badge/skills/kalinbogatzevski/captain-memo/stats.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00036 | $0.00731 |
| Opus 5.5 | $0.00014 | $0.00292 |
| Sonnet 5.5 | $0.00007 | $0.00146 |
| Haiku 4.5 | $0.00004 | $0.00073 |
Grade A, and why
stats 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 7d 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Captain Memo — corpus stats
When invoked, fetch the worker's stats and present them readably.
What to do
Call the captain-memo MCP tool stats (no arguments). It returns the worker's /stats body and sends the worker's secret for you: there is no token to handle. Where the CLI is installed, the same body is one command away, as one line of JSON:
captain-memo stats --json
If captain-memo is not on PATH, use the shim and checkout fallbacks described in the doctor skill.
The response includes: total_chunks, by_channel, observations (total/queue_pending/queue_processing), indexing (status/done/total/percent/elapsed_s/errors/last_error), project_id, embedder (model/endpoint), efficiency, and worker (started_at_epoch, uptime_s).
A successful /stats response means the backend worker is ONLINE — lead with that
and its uptime. Format worker.uptime_s compactly: 45s / 12m / 2h 13m / 3d 4h.
If the connection is refused or times out, the worker is OFFLINE: say so and point
the user at captain-memo doctor (see "On error"). Any other error (a 401, say) means the worker
answered: show the error text and point at captain-memo doctor.
Output format
Captain Memo — corpus statistics
────────────────────────────────
Worker: ● online · up 2h 13m
Project: <project_id>
Indexing: ready (or "indexing 47/275 (17%) ETA=3m 12s")
Total chunks: <total>
By channel:
memory <count>
skill <count>
capability <count>
observation <count>
Observations: <total> · <pending> pending · <processing> processing
Embedder: <model> @ <endpoint>
Efficiency
──────────
Compression: 16.4× — distilled 184,320 tokens of work into 11,240 stored
(94% saved · based on 312/340 observations)
Embedder: 47 calls · ~690 ms avg · 4,100 tok/s (since worker start)
Dedup: 95% of docs skipped re-embed (488/512 unchanged)
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.
- 7d ago Changed · +1 lines 97960eb54ff7
- 20d ago Changed · +5 lines d31381c3adfa
- 1mo ago First seen · 57 lines · 36 tokens per session scan A 3e0b68938d58
stats is a skill published in the GitHub repository kalinbogatzevski/captain-memo (5 stars, last pushed yesterday), licensed Apache-2.0. It adds 36 tokens to every session and 731 once invoked, about $0.0001 per session on Opus 5.5. 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
memem-mine
Opt in to memem's event-triggered mining. New sessions are mined automatically via the Stop hook.
curate
Power-user audit of Wenlan's pending surfaces: accept or dismiss pending revisions (conflicts/merges), or audit pending captures (not yet confirmed). Most users want /brief for revisions; that handles the daily flow. Use /wenlan curate only for explicit deep-walk audits after bulk imports, or to walk the full queue…
knowledge-base
Build and maintain a company knowledge base as a wiki of interlinked markdown notes in the workspace — a private, compounding Wikipedia. Use when the user wants to start or organize a knowledge base / wiki, ingest sources (URLs, documents, pasted notes) into it, ask questions answered from it, or audit (lint) it.…
pages
Browse, open, or print distilled pages, reusing the wenlan pages CLI for the actual open (one implementation on your PATH, no setup). /wenlan pages opens the page when one matches, or shows a small native picker when 2-4 match. /wenlan pages --print reads the page here instead of opening an editor. /wenlan pages lists…
forget
Delete a memory from Wenlan by ID. Destructive and cannot be undone — prefer /capture with supersedes for corrections. Invoked as /wenlan forget .
autosearch:context-retention-policy
Session-level policy for keeping the runtime AI's context window healthy across long research — keep-last-k tool results, offload older evidence to disk, trigger compaction at thresholds. Borrows MiroThinker's keeptoolresult, deepagents' summarization middleware, and deer-flow's SummarizationEvent pattern. Orthogonal…