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/muhdur/plsql-intelligence/agents-mdgit clone --depth 1 https://github.com/MuhDur/plsql-intelligenceWhat 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 | $0.03723 | $0.03723 |
| Opus 5 | $0.01861 | $0.01861 |
| Sonnet 5 | $0.00745 | $0.00745 |
| Haiku 4.5 | $0.00372 | $0.00372 |
Grade C, and why
plsql-intelligence 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 yesterday.
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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **C7 (operational, supersedes plan.md §3 literal text):** Agents may commit and push via the git skills below. Never run destructive ops (`git reset --hard`, `git clean -fd`, `git push --force`, branch deletion, `git s How it starts
The opening of the file, as written. The whole thing — 397 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — plsql-intelligence
Operating rules for agents working in this repository.
This is a greenfield Rust project now in active implementation. The working plan lives in plan.md; read it before doing anything substantive. If code and plan disagree, fix one or the other deliberately, never silently.
- Project key (for beads, agent mail, cass-memory):
plsql-intelligence - Repo name on disk:
oracle/(not the same as the project key) - Language: Rust (Cargo workspace). No JS/TS/Python in this repo.
- Status: Pre-GA implementation. Foundational crates exist; higher layers are still in progress.
- Web access: allowed. Use
WebFetch/WebSearchfor OSS docs, Oracle reference material, crate research, etc.
Founder constraints (plan.md §3, non-negotiable)
- C1/C2: No ML training on private data; no private code through founder-owned inference infra.
- C5/C6 + private estate: Private PL/SQL test material lives in the directory named by the
PLSQL_PRIVATE_ESTATEenvironment variable. Strictly local-only. Never publish, copy, orgit addanything from that path into this repo or any other git repository. Treat it as pattern reference only; test cases mirroring its patterns are re-synthesized from grammar + description, never copied. - C7 (operational, supersedes plan.md §3 literal text): Agents may commit and push via the git skills below. Never run destructive ops (
git reset --hard,git clean -fd,git push --force, branch deletion,git stash drop,rm -rfon tracked paths) without explicit in-session approval.
If unsure whether an action would violate one of these, stop and ask.
Plan rules to internalize (plan.md §4)
- R10 / R11: Every CLI ships
--robot-jsonand adoctorsubcommand. - R13: No uncertainty is silently dropped; every blind spot becomes a typed
UnknownReasonwith provenance. - R17: No telemetry by default.
- R20: Parser backend isolation is mandatory; no downstream crate depends on ANTLR-generated types.
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.
- yesterday First seen · 397 lines · 3,723 tokens per session scan C 4396006d91bd
plsql-intelligence AGENTS.md is an instructions file published in the GitHub repository MuhDur/plsql-intelligence (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 3,723 tokens to every session, about $0.0186 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
scip-query AGENTS.md
Instructions for PlunderStruck/scip-query, covering scip-query and git workflow.
scip-query CLAUDE.md
Instructions for PlunderStruck/scip-query, a project described as: Evidence and verification for AI coding agents: map code, reuse concepts, finish migrations, and gate diffs.
repobrain copilot-instructions.md
Copilot instructions for study8677/repobrain, covering github copilot bootstrap instructions and hard rule — query the repobrain hub first.
mirage CLAUDE.md
Instructions for strukto-ai/mirage, covering claude.md, repo layout, typescript packages, python/typescript parity and module layout.
mcp-steroid CLAUDE.md
Instructions for jonnyzzz/mcp-steroid, covering claude.md, agents.md, design philosophy, recursive context lookup (do this before sub-folder work), sub-folder guides and must do.
trace-mcp AGENTS.md
Instructions for nikolai-vysotskyi/trace-mcp: IMPORTANT: For ANY code exploration task, ALWAYS use trace-mcp tools first. NEVER use Read/Grep/Glob/Bash(ls,find) for navigating source code.