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/tenequm/pond/agents-mdgit clone --depth 1 https://github.com/tenequm/pondWhat 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 | $0.07770 | $0.07770 |
| Opus 5 | $0.03885 | $0.03885 |
| Sonnet 5 | $0.01554 | $0.01554 |
| Haiku 4.5 | $0.00777 | $0.00777 |
Grade A, and why
pond 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 3d 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 — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pond - project instructions
<comment_instructions>
- Minimalism: Write comments only when the why of the code is not obvious from the code itself. Never explain the what.
- Self-Documenting Code: Choose clear, descriptive names for variables, functions, and classes so that comments become redundant.
- Banned Comments: DO NOT include:
- Trivial explanations (e.g.,
# Loop through users). - "TODO" notes or author tags unless explicitly requested.
- Autogenerated file headers or boilerplate comments.
- Trivial explanations (e.g.,
- Formatting: Place comments on the same line as simple logic or directly above complex blocks. Keep explanations strictly under two sentences. </comment_instructions>
Commands
- Build:
cargo build - Test:
cargo test - Lint:
cargo clippy -- -D warnings - Format:
cargo fmt --check(usecargo fmtto fix) - Lockfile is enforced in CI with
--locked; locally, plain commands are fine. IfCargo.lockchanges unexpectedly,git statuswill show it.
Tests
PLACEMENT RULE - read before writing any test. A test lives next to the code it exercises. If it tests one module's behavior - even when it goes through a
Storeor handler call - it is a UNIT test and belongs in#[cfg(test)] mod testsat the bottom of thatsrc/...file.tests/integration/is ONLY for genuine cross-module integration: multiple subsystems wired end to end (e.g. ingest -> search -> get over a fixture corpus). AStore-level, single-module, or pure-helper test intests/integration/is in the WRONG place - move it tosrc/. When in doubt, default to a unit test insrc/.
- Layout (all crate paths are under
packages/pond/): unit tests live in#[cfg(test)] mod testsnext to the code they test (packages/pond/src/...). All integration tests are bundled into one binary atpackages/pond/tests/integration.rs, with each suite as a module underpackages/pond/tests/integration/<name>.rsand pulled in via#[path = ...] mod <name>;. Keep new integration suites in this folder and add a matching#[path]line - never drop a loosetests/foo.rsnext tointegration.rs(it would compile as a second binary and re-link the whole crate). - Adapter-specific integration tests go in
packages/pond/tests/integration/adapter/<adapter>.rs(mirroringpackages/pond/src/adapter/<adapter>.rs), not in concern-level files; keep cross-adapter interop (e.g. the foreign-restore matrix) inpackages/pond/tests/integration/adapter/mod.rs, the seam analog ofpackages/pond/src/adapter/mod.rs. - Run everything:
cargo test. - Run one integration suite:
cargo test --test integration -- <module>::(e.g.... -- search::). - Run one unit-test module:
cargo test --lib <module>::(e.g.... --lib sessions::tests::). - Run one test by name:
cargo test <name>(substring match across all binaries; add-- --exactto require a full match).
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.
- 3d ago First seen · 167 lines · 7,770 tokens per session scan A a1418e128fe0
pond AGENTS.md is an instructions file published in the GitHub repository tenequm/pond (57 stars, last pushed 4d ago), licensed Apache-2.0. It adds 7,770 tokens to every session, about $0.0389 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
quarry CLAUDE.md
Instructions for punt-labs/quarry, covering quarry, mandatory reading, architecture, how a query works and how ingestion works.
quarry AGENTS.md
Instructions for punt-labs/quarry, covering quarry for agents, what quarry does, integration model, mcp tools and slash commands.
quarry copilot-instructions.md
Instructions for punt-labs/quarry, covering copilot code review instructions, what to focus on, what not to flag, do not suggest and context.
quarry python.instructions.md
Instructions for punt-labs/quarry, covering python review rules, construction, encapsulation, types and error handling.
mcp-amr GEMINI.md
Instructions for AzureManagedRedis/mcp-amr, covering redis mcp server extension, what this extension provides, available tools, string operations and hash operations.
intervals-icu-mcp CLAUDE.md
Instructions for hhopke/intervals-icu-mcp, covering claude.md, project overview, development commands, architecture (quick reference) and tool categories.