AGENTS.md@src/ptc_agent/agent/prompts/templates agentsA
AGENTS.md@src/ptc_agent/agent/prompts/templates is agent-read markdown (agents) from ginlix-ai/langalpha.
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
# Prompt templates Jinja2 sources for the PTC, Flash, and subagent system prompts. Rendered by `PromptLoader` (`../loader.py`); defaults live in `../config/prompts.yaml`. Preview any surface as it renders at runtime: ```bash uv run python scripts/utils/render_prompt.py --mode ptc --crawl --count-tokens uv run python scripts/utils/render_prompt.py --subagent equity-analyst ``` ## Two guidance levels: lean and detailed The service is model-agnostic — the same prompt has to serve frontier models and small self-hosted ones. Frontier models regress when given procedural scaffolding they already fold in; a 20b model falls apart without it. So every template renders at one of two levels, resolved per request (see `../guidance.py`): ``` user model_preference.prompt_guidance → config.yaml prompt.guidance → models.json prompt_guidance → detailed ``` `detailed` is the fail-safe default: an unannotated or unknown model gets more scaffolding, never less. ### You never write the lean version There is one template body. **Lean is what remains when the coaching is stripped** — it is not a second text to keep in sync. The split is: | | Goes where | Examples | |---|---|---| …
Read the whole file at its exact version.
How to install
mdr add ginlix-ai/langalpha/src/ptc_agent/agent/prompts/templates/AGENTS.md@git:20260913.8fae40dmdr add ginlix-ai/langalpha/src/ptc_agent/agent/prompts/templates/AGENTS.md@sha256:1e66ec5a31a0ba5fPin 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_3otkfyekyzrmk4o3)
1 badge views in 30 days
Versions
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260913.8fae40d latest | 2026-09-13 | 8fae40d | 9,849 B | A | view · diff |
| git:20260901.8abca61 | 2026-09-01 | 8abca61 | 6,244 B | A | view |
Audit of the latest version
- pass: Size between 200 bytes and 200 KB (9849 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
ginlix-ai/langalpha · 1,774 stars · license Apache-2.0 · pushed 2026-09-24 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_3otkfyekyzrmk4o3 GET https://markdownregistry.com/api/v1/resolve?ref=ginlix-ai/langalpha/src/ptc_agent/agent/prompts/templates/AGENTS.md GET https://markdownregistry.com/api/v1/blob/1e66ec5a31a0ba5f882c70f22c6edd195761b59eef5ff624c011d1222598cecd
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.