Borrowing it
Nothing to install: this file belongs to mmnto-ai/totem. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/mmnto-ai/totem/main/.agents/skills/signoff/SKILL.mdgit clone --depth 1 https://github.com/mmnto-ai/totemWrote 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/mmnto-ai/totem/signoff)<a href="https://agentmods.dev/skills/mmnto-ai/totem/signoff"><img src="https://agentmods.dev/badge/skills/mmnto-ai/totem/signoff/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/mmnto-ai/totem/signoff"><img src="https://agentmods.dev/badge/skills/mmnto-ai/totem/signoff.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 6 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
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 | $0.00016 | $0.02590 |
| Opus 5 | $0.00008 | $0.01295 |
| Sonnet 5 | $0.00003 | $0.00518 |
| Haiku 4.5 | $0.00002 | $0.00259 |
Grade A, and why
signoff 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 11d 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
End-of-session wrap-up. Post-Proposal-282 (ADR-106), journals + handoffs live in the per-repo .totem/orchestration/<agent-id>/ tree (gitignored) — NOT the substrate. Substrate stays as a frozen archive for forensic reads; the active surface is local.
-
Crown gate — derive FIRST (Prop 305 §8; operator-ruled 2026-08-13; non-cohort scope amended by operator word 2026-08-14, recorded on the mmnto-ai/totem-strategy#488 R-table). Journal rule first — every seat, every repo: restate your OWN chain's still-open
owed:lines only — lines where YOU are the waiter and@holder:names the counterparty who owes (naming yourself as@holder:makes a note-to-self the parser silently skips) — and never another seat's carryforward: the parser keys each edge's waiter to the journal-owning seat, so a copied line mints a live foreign edge in YOUR chain (a line self-held at its origin is a skipped note-to-self THERE and a LIVE edge in yours) that persists until your own next journal drops it. (Canonical rulings home: mmnto-ai/totem-status#127.) Then the gate ladder: read thetotem:agent-busmarker in this repo's AGENTS.md and check the step-2a cohort table below. No declared marker (absent, or missingrole/seat— the parity sensor's not-declared set) AND this repo's basename is NOT in the cohort table = non-cohort consumer: the crown gate does not apply — run your full signoff. No declared marker but the repo IS in the table = a cohort repo that has not normalized: treat as crown vacant (fail-closed — the ruled default). Marker declared: deriveprimary=fresh — that seat is the crown; declared but noprimary== crown vacant. Vacant (either arm) is a legitimate state: surface upkeep stages/defers, mail the operator, and no seat self-promotes (transfer is the operator's word, Prop 305 §8.1; vacancy legitimacy §8.4b). Before comparing yourself toprimary=, verify your OWN identity — the signon S0 check applies at signoff too: your seat derives per step 2a, and a mis-scoped or inheritedTOTEM_SELF_AGENTnaming a foreign seat must not make this session the crown; a foreign or ambiguous self-identity is fail-closed — no crown decision, no upkeep mutations — until the identity is fixed. If you are NOT the crown, your signoff is seat-anchored + own-lane: skip the surface-upkeep MUTATIONS — derived-cache/corpus rebuilds, GH board writes, cohort-wide upkeep walks, and step 4's shared-tree branch cleanup (crown-scoped there by name) — while every OTHER managed step (1–3, 5–6) still runs for every seat, and read-only derivation stays open to all: derive, then report staleness to the crown by mail. -
Update memory. Update auto-memory files (e.g.
MEMORY.md, topic memories) with any new state — version shipped, tickets closed, key decisions, banked feedback or doctrine signals. -
Write a journal entry to the per-repo orchestration path. Filename convention:
<model>-NNNN-<short-topic-slug>.md(e.g.,claude-0057-phase-4-resolver-shipped.md).Resolve the path two steps:
a. Identify your agent-id from the current repo's basename. The hardcoded map (Proposal 282 § Scope item 3 — keep in sync with the ADR-106 cohort list):
Repo ( git rev-parse --show-toplevelbasename)Claude agent-id Gemini agent-id Kimi agent-id totemtotem-claudetotem-geminitotem-kimitotem-strategystrategy-claudestrategy-gemini(not seated) liquid-citylc-claudelc-gemini(not seated) arhgap11arhgap11-claudearhgap11-gemini(not seated) totem-statusstatus-claudestatus-gemini(not seated) totem-playground(orphan stream — no native agent) (orphan stream) (orphan stream)
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.
- 11d ago First seen · 56 lines · 16 tokens per session scan A 722011a9f6b0
signoff is a skill published in the GitHub repository mmnto-ai/totem (17 stars, last pushed yesterday), licensed Apache-2.0. It adds 16 tokens to every session and 2,590 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
feature-knowledge
Structures codebase exploration into a feature knowledge base.
performance
This skill should be used when reviewing code for N+1 queries, memory leaks, or I/O bottlenecks.
rust
This skill should be used when the user works with Rust files (.rs), asks about "ownership", "borrowing", "lifetimes", "Result/Option", "traits", or discusses memory safety and type-driven design. Provides patterns for ownership, error handling, type system usage, and safe concurrency.
setup-obsidian
Turn a repo's docs folder into a retrieval graph — router, generated indexes, state file — and open it as an Obsidian vault.
review
Validate plans, execution, or PRs against wish criteria — returns SHIP / FIX-FIRST / BLOCKED with severity-tagged gaps.
learn
Diagnose and fix agent behavioral surfaces when the user corrects a mistake — connects to Claude native memory.