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 xg-gh-25/SwarmAI --skill s_ddd-persistgit clone --depth 1 https://github.com/xg-gh-25/SwarmAIWrote 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/xg-gh-25/swarmai/s_ddd-persist)<a href="https://agentmods.dev/skills/xg-gh-25/swarmai/s_ddd-persist"><img src="https://agentmods.dev/badge/skills/xg-gh-25/swarmai/s_ddd-persist/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/xg-gh-25/swarmai/s_ddd-persist"><img src="https://agentmods.dev/badge/skills/xg-gh-25/swarmai/s_ddd-persist.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, 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 82 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.
- medium Memory Poisoning · line 50 Skill injects content designed to persist in agent memory or context across interactions. Persistent injection can alter agent behavior long after the initial interaction.Fix: Do not allow untrusted input to persist in agent memory or context. Validate all content before storing and implement memory isolation between sessions.
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 | $0.00154 | $0.02037 |
| Opus 5 | $0.00077 | $0.01019 |
| Sonnet 5 | $0.00031 | $0.00407 |
| Haiku 4.5 | $0.00015 | $0.00204 |
Grade A, and why
ddd-persist 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 9d 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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DDD Persist (s_ddd-persist) — sediment cognition into the DDD, with discipline
Route knowledge into the correct section of THIS DDD, additively, under a lock. This is the WRITE side of the self-養成 loop: what a dev-loop's REFLECT stage learns (whatever pipeline the host runtime runs over this DDD) gets sedimented here so the next run is smarter. But persisting is a judgment, not a reflex — most passing facts should be SKIPPED (see Step 0).
DDD-native decouple of SwarmAI's
s_persist. Same routing discipline + admission philosophy, re-homed to the DDD's own files. The locked read-modify-write engine ships bundled (scripts/locked_write.py, pure-stdlib + fcntl); the SwarmAI injection-guards (core.memory_guard) are fail-soft absent — a DDD's docs aren't SwarmAI's always-injected MEMORY.md. Nodata.db, no SwarmAI memory index, no backend.
The Routing Decision Tree (3 steps — do them in order)
Step 0: Should this be persisted AT ALL? (admission gate — default is SKIP)
Persisting is not free. A stored fact that drifts becomes a liability: stale → it misleads; fresh → it costs upkeep; and if it never drove a judgment, it was worthless the moment it was written. Most passing facts should NOT be persisted.
REJECT (do not persist) — volatile, zero-decision-value data:
- Counts that drift: LOC, file/test counts, line numbers, star/fork snapshots, token sizes, "N sessions / N commits", percentages-of-the-moment.
- A raw number whose only use is "to know the number." If it doesn't change a future decision, it's a measurement — take measurements live, don't store them.
- Status true only right now: "build is green", "3 tabs open".
- One-off transient context with no cross-session reuse.
For a drifting-but-occasionally-needed metric: store the reproducible method (the command that regenerates it), NEVER the frozen output. Or describe it qualitatively.
ADMIT (persist) only if BOTH hold:
- Decision-relevant — it will change how the agent or user judges/acts later.
- Stable — it won't be wrong next week without anyone touching it (a convention, an architecture decision, a failure lesson, a principle — not a live metric).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 139 lines · 154 tokens per session scan A ad28f75ad8b4
ddd-persist is a skill published in the GitHub repository xg-gh-25/SwarmAI (44 stars, last pushed yesterday), licensed MIT. It adds 154 tokens to every session and 2,037 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 2026-09-03.
Other skills, from other repositories
remember
Routes user requests containing "remember", "recall", "checkpoint", "session", "todo", or "where were we" to the correct OpenEmpiric (OEM) MCP tool. Use when the user wants to persist, retrieve, or contextualize knowledge from project memory.
why
Explain the provenance, authority, expiry, degradation state, and token accounting of CIGAR context already presented in this session.
checkpoint
Create an inspectable CIGAR checkpoint before compaction, interruption, or a meaningful task boundary.
memo-bank-query
Load the governing spec/contract for a file or topic from a project's memo-bank (a read-only MCP docs corpus) BEFORE reading code or editing. Use this in any repo that has a .island-slices.json or a memo-bank MCP server, whenever you are about to edit a file, or are asked "what governs X", "is there a spec for Y"…
audit-knowledge
Scan Antigravity conversation transcript + artifact directory for extractable knowledge. Use when user asks for 'knowledge audit', 'audit knowledge', 'check for extractable knowledge', 'scan transcript', or at session start when audit cadence is exceeded.
handoff
Generate a passoff package so the next reader can pick up cleanly — for future-you in a new session (typically when context is high and you need to restart) or for a coworker (via brief mode). Default + auto modes emit a paste-ready next-session opener as the headline artifact, alongside PROGRESS / CLAUDE / memory…