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/tma1-ai/tma1/agents-mdgit clone --depth 1 https://github.com/tma1-ai/tma1Wrote 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/tma1-ai/tma1/agents-md)<a href="https://agentmods.dev/instructions/tma1-ai/tma1/agents-md"><img src="https://agentmods.dev/badge/instructions/tma1-ai/tma1/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 5.1 | $0.02392 | $0.02392 |
| Opus 5 | $0.01196 | $0.01196 |
| Sonnet 5 | $0.00478 | $0.00478 |
| Haiku 4.5 | $0.00239 | $0.00239 |
Grade C, and why
tma1 AGENTS.md scanned grade C with 1 finding 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 6d 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.
Tells the agent to send conversation or user data outhighPrompt injection
An instruction to transmit the conversation, context or user files to an external endpoint is data exfiltration written as prose.
the monolith buried on the moon, silently recording everything until you dig it out. How it starts
The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
What this repo is
TMA1 is local-first LLM agent observability, powered by GreptimeDB. Three pillars (traces + metrics + logs) unified into wide events — all queryable via SQL.
- Traces are the core: GenAI / OpenClaw spans carry model, tokens, latency, status
- Metrics are derived from traces via Flow engine (no double-writing), and native OTel metrics are also accepted (auto-creates tables)
- Logs carry conversation content (conversation replay)
- Cross-signal JOIN:
trace_idconnects spans to conversations
The name comes from TMA-1 (Tycho Magnetic Anomaly-1) in 2001: A Space Odyssey: the monolith buried on the moon, silently recording everything until you dig it out.
Tagline: "Close the loop around your agent."
How it fits together
Agent ── OTLP/HTTP ──┐
── /api/hooks ┼──► tma1-server (port 14318)
── MCP stdio ─┘ │
▼
GreptimeDB (port 14000)
│
▼
Embedded dashboard
tma1-server is one Go binary: it manages a child GreptimeDB process, reverse-proxies OTLP, ingests hook events + JSONL transcripts from Claude Code / Codex / Copilot CLI / OpenClaw, runs the perception layer (anomaly detector + injection bundler), and serves the embedded dashboard. The MCP child (tma1-server mcp-serve) shares the same GreptimeDB.
Deep references — read these when you need detail:
docs/architecture.md— module layout, full data flow, per-agent tables, env vars, file indexdocs/hooks.md— hook injection protocol (5 events × 2 adapters)docs/mcp-tools.md— 10 MCP tools backed by the perception bundlerdocs/anomalies.md— anomaly rules, channels, suppression
Commands
make build # Build the binary (sync-plugin first so embedded skill/command tree is fresh)
make build-linux # Cross-compile for Linux amd64
make build-windows # Cross-compile for Windows amd64
make run # Build and run locally
make dev # Auto-rebuild + restart on .go / .html / .css / .js / .sql change
make install # Install dev build to $(INSTALL_DIR) — default ~/.tma1/bin — without restarting the running server
make sync-plugin # Manually mirror claude-plugin/{skills,commands} → server/internal/hooks/{skills,commands} for go:embed
make vet # go vet on ./cmd/... ./internal/... ./web
make lint # golangci-lint v2 on the same selector
make lint-js # ESLint on dashboard JS (requires Node.js)
make test # Tests with race detector on the same selector
# CI also runs: golangci-lint + ESLint + shellcheck site/public/install.sh + PSScriptAnalyzer on install.ps1
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.
- 6d ago First seen · 153 lines · 2,392 tokens per session scan C ea8e3fca3935
tma1 AGENTS.md is an instructions file published in the GitHub repository tma1-ai/tma1 (117 stars, last pushed 2d ago), licensed Apache-2.0. It adds 2,392 tokens to every session, about $0.0120 per session on Opus 5. A static security scan graded it C with 1 finding (tells the agent to send conversation or user data out). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
clawmetry AGENTS.md
AGENTS.md instructions for vivekchand/clawmetry, covering agents.md — for ai coding agents, quick context, where new code goes (open-core split), the rules that bite and common tasks.
FlowBoard AGENTS.md
Instructions for rasimme/FlowBoard, covering flowboard — agent entry point, start here and where things live.
ClawBio AGENTS.md
AGENTS.md instructions for ClawBio/ClawBio, covering agents.md — clawbio guide for ai coding agents, project overview, setup, clone and install dependencies.
agiwo AGENTS.md
Instructions for xhwSkhizein/agiwo, covering agents.md, how to use this file, truth source order (adr 0048 / 0049), logging and config hot reload.
clawmetry CLAUDE.md
Claude Code instructions for vivekchand/clawmetry, covering claude.md — clawmetry, what is this?, architecture, key files and core.
ClawBio CLAUDE.md
Claude Code instructions for ClawBio/ClawBio, covering claude.md — clawbio agent instructions, homepage (index.html) — strict, key files, slash commands and skill routing table.