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/driftya/code-meridian/agents-mdgit clone --depth 1 https://github.com/Driftya/code-meridianWhat 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 | $0.00915 | $0.00915 |
| Opus 5 | $0.00458 | $0.00458 |
| Sonnet 5 | $0.00183 | $0.00183 |
| Haiku 4.5 | $0.00092 | $0.00092 |
Grade A, and why
code-meridian 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 — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Short always-on guidance for automated agents working in this repository.
Read this first. Use the detailed references under docs/agent/ when the task needs more depth.
Behavior Expectations
- Do not assume silently. State important assumptions.
- If something is unclear, ask instead of guessing.
- Prefer the simplest solution that solves the request.
- Do not add speculative abstractions, flexibility, or features.
- Make surgical changes only. Avoid unrelated refactors.
- Match the existing style unless the task requires otherwise.
- Remove only unused code introduced by your own change unless asked.
- Define success criteria and verify them.
- For bugs, prefer reproduce -> fix -> verify.
- Every changed line should trace back to the request.
Required CodeMeridian Usage
Use CodeMeridian proactively. Prefer graph tools over terminal scans when the graph can answer the question.
Trigger rules
| Situation | Tool to call |
|---|---|
| Before any non-trivial edit | build_minimal_context |
| Before editing a specific method/class | resolve_exact_symbol, then get_context_for_editing |
| Before a refactor | find_impact and find_test_shield |
| Before deleting code | find_unreferenced |
| Before starting a feature | analyze_feature_implementation_path, then find_implementation_surface for exact targets |
| Before trusting exact file targets | check_graph_freshness or find_graph_drift |
| "How do X and Y relate?" | find_connection |
| Frontend HTML/CSS/SCSS relationship work | build_minimal_context, then find_connection or find_implementation_surface |
| Before renaming or deleting a frontend class/selector/style import | find_impact |
| Investigating CSS/SCSS cascade or specificity issues | find_frontend_cascade_conflicts |
| Looking for repeated style values or token extraction candidates | find_duplicate_candidates |
| Looking for an existing implementation shape to mirror | find_implementation_patterns |
| Looking for duplicate/refactor risk | find_duplicate_candidates or find_similar_nodes |
| Looking for missing tests | find_coverage_gaps |
| Searching docs/decisions | search_documentation |
| Working with config | find_config_definitions and find_config_usage |
| Looking for keyword-related context | find_related_knowledge |
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 · 90 lines · 915 tokens per session scan A ab8401926d1e
code-meridian AGENTS.md is an instructions file published in the GitHub repository Driftya/code-meridian (4 stars, last pushed 3d ago), licensed MIT. It adds 915 tokens to every session, about $0.0046 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 2026-08-31.
Other instructions, from other repositories
CodeNib AGENTS.md
Instructions for sysevol-ai/CodeNib, covering agents.md, layered objectives, dev commands, git, commit, and pr rules and testing guidance.
repobrain copilot-instructions.md
Copilot instructions for study8677/repobrain, covering github copilot bootstrap instructions and hard rule — query the repobrain hub first.
repobrain AGENTS.md
AGENTS.md instructions for study8677/repobrain: Beautiful is better than ugly. Explicit is better than implicit. Simple is better than complex. Complex is better than complicated. Flat is better than nested. Sparse is better than dense. Readability counts. Special cases aren't special enough to break the rules.…
sverklo CLAUDE.md
Instructions for sverklo/sverklo, covering sverklo development, using sverklo tools, project structure and build & test.
sourcebot CLAUDE.md
Claude Code instructions for sourcebot-dev/sourcebot, covering claude code guidelines, database migrations, building packages, backend workloads and execution locks.
codesage AGENTS.md
Instructions for iliaal/codesage, covering codesage, build, sanity check before pushing, crate map and search pipeline.