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 Zhonghao1995/agentic-swmm-workflow --skill swmm-rag-memorygit clone --depth 1 https://github.com/Zhonghao1995/agentic-swmm-workflowWrote 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/zhonghao1995/agentic-swmm-workflow/swmm-rag-memory)<a href="https://agentmods.dev/skills/zhonghao1995/agentic-swmm-workflow/swmm-rag-memory"><img src="https://agentmods.dev/badge/skills/zhonghao1995/agentic-swmm-workflow/swmm-rag-memory/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/zhonghao1995/agentic-swmm-workflow/swmm-rag-memory"><img src="https://agentmods.dev/badge/skills/zhonghao1995/agentic-swmm-workflow/swmm-rag-memory.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 Rogue Agent · line 19 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
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 | $0.00062 | $0.01209 |
| Opus 5 | $0.00031 | $0.00605 |
| Sonnet 5 | $0.00012 | $0.00242 |
| Haiku 4.5 | $0.00006 | $0.00121 |
Grade A, and why
swmm-rag-memory 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 10d 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 — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SWMM RAG Memory
What this skill provides
- Query-time retrieval over Agentic SWMM audited run memory.
- A lightweight keyword/tag retriever that works without embeddings or a vector database.
- A local hybrid retriever that combines keyword matches, deterministic SWMM tags, metadata weighting, and hashed token/character n-gram embeddings.
- RAG context packs that can be passed to Codex, OpenClaw, Hermes, or another LLM.
- Source citations for each retrieved memory item, including run id, project key, source file, failure patterns, diagnostics, and matched terms.
- Retrieval-grounded
failure_advice.{json,md}for failed or warning runs, without modifying model files. - Explicit
resolution_memory.jsonfor human-reviewed and benchmark-verified repairs. - Obsidian-compatible Markdown output for saved retrieval notes.
This skill reads existing audit and modeling-memory artifacts. It does not run SWMM, modify model inputs, rewrite skills, or claim that retrieved memory proves a modeling conclusion.
Relationship to swmm-modeling-memory
swmm-modeling-memory summarizes audited runs after experiments have been recorded.
swmm-rag-memory retrieves the most relevant historical memory for a current question.
The intended loop is:
- Run SWMM or attempt a workflow.
- Audit the run.
- Refresh
swmm-modeling-memory. - Ask a current modeling question.
- Retrieve relevant historical memory with
swmm-rag-memory. - Answer with explicit source boundaries and citations.
Output contract
The corpus builder writes these files to the selected RAG-memory output directory:
corpus.jsonlkeyword_index.jsonembedding_index.json
The retriever writes JSON results by default and can also write a Markdown context pack. Failure advice writes failure_advice.json and failure_advice.md into the run directory. Verified repairs can be recorded as resolution_memory.json.
CLI
Build a corpus from existing memory and audited runs:
python3 skills/swmm-rag-memory/scripts/build_memory_corpus.py \
--memory-dir memory/modeling-memory \
--runs-dir runs \
--out-dir memory/rag-memory
What ships with it
7 files 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.
- scripts/answer_with_memory.py 2.0 KB runs code
- scripts/build_memory_corpus.py 14 KB runs code
- scripts/generate_failure_advice.py 4.1 KB runs code
- scripts/rag_memory_lib.py 35 KB runs code
- scripts/record_resolution_memory.py 4.0 KB runs code
- scripts/refresh_after_run.py 5.1 KB runs code
- scripts/retrieve_memory.py 2.2 KB runs code
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.
- 10d ago First seen · 142 lines · 62 tokens per session scan A 68962c70736a
swmm-rag-memory is a skill published in the GitHub repository Zhonghao1995/agentic-swmm-workflow (27 stars, last pushed 3d ago), licensed MIT. It adds 62 tokens to every session and 1,209 once invoked, about $0.0003 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
LangChain RAG Pipeline
INVOKE THIS SKILL when building ANY retrieval-augmented generation (RAG) system. Covers document loaders, RecursiveCharacterTextSplitter, embeddings (OpenAI), and vector stores (Chroma, FAISS, Pinecone).
graphify
Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community…
digital-health-clinical-asr-finetune
Stage 4 of the Clinical ASR Flywheel. Use when priority KER is above 0.3 to run stock NeMo SFT on Parakeet TDT v2 and offline cycle N+1 re-eval. NOT for generic word boosting (use /finetune-asr).
digital-health-clinical-asr-eval
Stage 3 of Clinical ASR Flywheel. Score a NeMo manifest, produce the five-section KER leaderboard (by-ipasource diagnostic). Not for ASR auth (/riva-asr).
nemo-mbridge-perf-expert-parallel-overlap
Validate and use MoE expert-parallel communication overlap in Megatron-Bridge, including overlapmoeexpertparallelcomm, delaywgradcompute, and flex dispatcher backends such as DeepEP and HybridEP.
jetson-inference-mem-tune
Pick the serving stack and per-runtime memory flags (vLLM, SGLang, llama.cpp, TensorRT Edge-LLM) for an LLM/VLM workload on any NVIDIA Jetson.