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 agentmods add instructions/flop-labs/technocore-chat/agents-mdgit clone --depth 1 https://github.com/flop-labs/technocore-chatWrote 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/instructions/flop-labs/technocore-chat/agents-md)<a href="https://agentmods.dev/instructions/flop-labs/technocore-chat/agents-md"><img src="https://agentmods.dev/badge/instructions/flop-labs/technocore-chat/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00621 | $0.00621 |
| Opus 5 | $0.00311 | $0.00311 |
| Sonnet 5 | $0.00124 | $0.00124 |
| Haiku 4.5 | $0.00062 | $0.00062 |
Grade A, and why
technocore-chat AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
AGENTS.md
CI runs exactly these — run them before pushing:
uv sync --frozen
uv run ruff check .
uv run ruff format --check .
uv run ty check
uv run coverage run -m pytest tests -q
uv run coverage report
Layering: the core is src/store.py, src/didkey.py, src/config.py, src/limit.py
and a thin src/app.py adapter.
src/manifest.py, docs and frontends are extra — never counted as core.
Core size caps live in sz-baseline.json (caps); check with uv run sz.py --caps.
Growth past a cap needs a new primitive, or belongs in extra.
-
If uv cannot write its default cache (sandboxed agent/CI environments), use
UV_CACHE_DIR=$PWD/.uvcache— a worktree-local cache always works. -
Input handling has one rule, written down once: docs/design.md §3.5 — advisory params (
limit,wait,n,format,since) clamp and the schema documents the clamp; semantic ones (identity, content,if=/if_absent, names) refuse with a 400 naming the field.tests/test_contract.pyfails the build on drift either way. -
File against HEAD, verified on a fresh fetch — main moves several times a day, and a snapshot review reports findings against code that no longer exists. Name the commit; the issue form requires it.
-
One issue, one PR: if an open PR already covers it, review or extend that one. CI comments on a new PR when open PRs cite the same issue (queue-guard.yml).
-
Never edit CHANGELOG.md or sz-baseline.json in an ordinary PR — both are maintainer-regenerated. Release-note wording goes in the PR body; if a size cap binds, propose the missing primitive or move the change to extra. A release PR a maintainer asked for is the exception: folding [Unreleased] into a dated section is the packaging, and release.yml refuses to tag a version CHANGELOG.md has no section for. queue-guard only fails fork PRs, so on an in-repo branch this is a rule, not a check.
-
Move tests, don't rewrite them: test bodies stay byte-identical across reorganisations.
-
Size the core with
uv run sz.py(table) anduv run sz.py --check(fails if core code-lines grew pastsz-baseline.json). Its PEP 723 header makes it standalone:uv runit from anywhere, no project env needed. -
quint-specs/, quint-oracle.*.json, src/oracle_client.py, tests/conftest.py, preflight.sh, submit.sh are local-only WIP — never commit them.
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.
- today Changed · +3 lines · +79 tokens per session db2a07dd2c94
- 4d ago First seen · 41 lines · 542 tokens per session scan A 10f69ee14bfc
technocore-chat AGENTS.md is an instructions file published in the GitHub repository flop-labs/technocore-chat (146 stars, last pushed today), licensed Apache-2.0. It adds 621 tokens to every session, about $0.0031 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 instructions, from other repositories
technocore-chat AGENTS.md
AGENTS.md instructions for flop-labs-dev/technocore-chat: CI runs exactly these — run them before pushing.
AionUi AGENTS.md
AGENTS.md instructions for iOfficeAI/AionUi, covering aionui - project guide, code conventions, file & directory structure, naming and ui library & icons.
fatecat AGENTS.md
Instructions for tradecatlabs/fatecat, covering agents.md - fatecat enterprise repo, 目录用途, 目录结构, 职责边界 and web html 设计硬规则.
Risuai AGENTS.md
AGENTS.md instructions for kwaroran/Risuai, covering project overview, directory structure, source code structure (/src), building and running and prerequisites.
tool-ui CLAUDE.md
Claude Code instructions for assistant-ui/tool-ui, covering tool ui, commands, tooling, stack and architecture.
agent-skills AGENTS.md
AGENTS.md instructions for GetStream/agent-skills, covering agents.md - codex entrypoint for the stream skill pack, sub-skills, codex-specific and install.