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/simonsysun/seeklink/agents-mdgit clone --depth 1 https://github.com/simonsysun/seeklinkWhat 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 | $0.01484 | $0.01484 |
| Opus 5 | $0.00742 | $0.00742 |
| Sonnet 5 | $0.00297 | $0.00297 |
| Haiku 4.5 | $0.00148 | $0.00148 |
Grade A, and why
seeklink 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 2d 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 — 237 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project conventions for AI coding agents
Core model
The source of truth is the repository and git history.
Sessions, chat context, local notes, and worktrees are convenience layers only. Important state must be recoverable from repo files or git history.
Keep this file short. Add rules only when they affect agent behavior or prevent repeated mistakes.
Start of session
Before editing:
- Read
docs/current_focus.mdif it exists. - Check branch and git status.
- Identify relevant verification commands.
- For non-trivial changes, state the approach before writing code.
Do not rely on chat history as project memory.
Current focus
Use docs/current_focus.md for current work state only. Keep it under 30 lines.
It should contain:
- current task;
- branch;
- next concrete step;
- open questions;
- verification commands;
- at most the latest two handoff notes.
Rewrite stale content instead of appending indefinitely.
Handoff format:
Handoff (YYYY-MM-DD, Agent, branch)
Done: Open: Next: Status: <in progress / blocked / ready for review>
Do not create changelog, backlog, task tracker, memory, or per-agent state files unless explicitly requested.
History and decisions
Commit messages are the primary history layer.
Use commit bodies to explain why non-trivial changes were made.
Use docs/decisions.md only for decisions that remain active constraints on
future work. Do not create it by default.
Decision docs are appropriate when a decision:
- spans multiple commits or subsystems;
- involves external context not visible in code;
- prevents a recurring wrong choice;
- affects public API, data model, deployment, security, or product behavior.
Do not mirror every commit into decision docs.
Agents may propose decision entries, but must not create or update long-lived decision records unless the human approves the decision text.
If docs/decisions.md grows beyond roughly 10 active decisions, propose moving
to docs/decisions/YYYY-MM-DD-<slug>.md ADR files.
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.
- 2d ago First seen · 237 lines · 1,484 tokens per session scan A c56361cb4a80
seeklink AGENTS.md is an instructions file published in the GitHub repository simonsysun/seeklink (11 stars, last pushed 2mo ago), licensed MIT. It adds 1,484 tokens to every session, about $0.0074 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
mind-mem CLAUDE.md
Instructions for star-ga/mind-mem, covering mind-mem — persistent ai memory system, architecture, key components, mcp tools (97) and config.
mind-mem AGENTS.md
Instructions for star-ga/mind-mem, covering mind-mem: agent instructions (auto-written), memory context, apr 21, 2026, may 2, 2026 and may 6, 2026.
mind-mem copilot-instructions.md
Instructions for star-ga/mind-mem: This repository uses mind-mem for persistent memory. Before answering, consult memory with mm inject --agent copilot. Respect ADR / DECISION blocks; route new decisions through proposeupdate rather than modifying them directly.
memex AGENTS.md
Instructions for nicosuave/memex, covering agents and before committing.
nexusos AGENTS.md
Instructions for asimons81/nexusos, covering agents.md, mission, isolation rules, architecture boundary and current release boundary.
Connapse CLAUDE.md
Instructions for Destrayon/Connapse, covering claude.md, build & run, build, run web app (requires backing services) and full stack via docker.