agent-graph skillA
agent-graph is agent-read markdown (skill) from badmuriss/my-llm-kit: Run durable task graphs when independent work needs ownership, integration checks and recovery. Use for graph execution, not routine edits..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Agent Graph Use the durable graph only when the decision proves independent packets, bounded ownership, checks, integration, budget and cleanup. Ordinary planning and implementation do not require this runtime. Resolve this installed skill directory and invoke its `scripts/agent_graph.py` with `--repo <absolute-project-path>`, using `python3` or Windows `py -3`. The Python environment needs `jsonschema`; see `requirements.txt`. ## Adaptive entry `agent_graph.py intake` returns `direct`, `verified_single`, `light_spec`, or `graph`. It reads bounded repository facts before asking a question. It creates no run or worker. - Direct and verified-single work need no OpenSpec, graph, Canvas, or Orca process. - Light spec uses one amendable Markdown decision record. OpenSpec remains optional. - Graph mode alone creates the durable journal and coordinator claim. Persist a graph selection as `openspec/changes/<slug>/process-decision.json`. Bootstrap validates that its current decision revision, packet paths, checks, permission, budget, integrator, and cleanup plan match the OpenSpec task graph before creating run artifacts. …
Read the whole file at its exact version.
How to install
mdr add badmuriss/my-llm-kit/agent-graph@git:20260921.471bbedmdr add badmuriss/my-llm-kit/agent-graph@sha256:0a3421c81f9fa3a8Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_ebh6uski6isgfphr)
1 badge views in 30 days
Versions
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260921.471bbed latest | 2026-09-21 | 471bbed | 6,463 B | A | view · diff |
| git:20260906.3d137c0 | 2026-09-06 | 3d137c0 | 6,302 B | A | view · diff |
| git:20260826.7eae05d | 2026-08-26 | 7eae05d | 6,253 B | A | view · diff |
| git:20260826.22b176c | 2026-08-26 | 22b176c | 5,202 B | A | view · diff |
| git:20260820.f989036 | 2026-08-20 | f989036 | 1,805 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (6463 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
badmuriss/my-llm-kit · 3 stars · license MIT · pushed 2026-09-23 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_ebh6uski6isgfphr GET https://markdownregistry.com/api/v1/resolve?ref=badmuriss/my-llm-kit/agent-graph GET https://markdownregistry.com/api/v1/blob/0a3421c81f9fa3a8b9dce310dd31c43de06318f0bd871ff2ddc272260d2aa5f8
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.