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/darrenbenson/sdlc-studio/agents-mdgit clone --depth 1 https://github.com/DarrenBenson/sdlc-studioWrote 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/darrenbenson/sdlc-studio/agents-md)<a href="https://agentmods.dev/instructions/darrenbenson/sdlc-studio/agents-md"><img src="https://agentmods.dev/badge/instructions/darrenbenson/sdlc-studio/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 | $0.03662 | $0.03662 |
| Opus 5 | $0.01831 | $0.01831 |
| Sonnet 5 | $0.00732 | $0.00732 |
| Haiku 4.5 | $0.00366 | $0.00366 |
Grade A, and why
sdlc-studio AGENTS.md scanned grade A 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 4d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| `reconcile --verify` | whichever of `pytest`, `jest`, `vitest`, `go`, `curl`, `jq`, `rg` your Verify lines invoke | How it starts
The opening of the file, as written. The whole thing — 252 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for coding agents working on this repository. Claude Code reads this via the
@AGENTS.md import in CLAUDE.md; Codex, Copilot, Cursor, Gemini and others
read it directly.
This file is always loaded. Keep it short enough to be read, and put nothing here that a command already prints. The pre-commit hook names every guard it runs, the rule it enforces, the offending line and the fix, so this file does not restate them.
Start here, every session
Do these four before acting, including after a context reset or compaction. They take under a minute and each one has cost a real sprint when skipped.
sdlc-studio/reviews/LATEST.md- what the last run landed and what it left owed./sdlc-studio status- pipeline state and the next step..claude/skills/sdlc-studio/lessons/_index.md- the cross-project lessons. Read them before any design decision, not after.git fetch- a stale clone mints ids the remote already used and plans over artefacts somebody else changed.
The failure mode this repo actually has
Not ignorance. Rules that were read and then not followed. One session produced five
instances: the lessons index went unread while the plan printed its digest; review prompts
were hand-written though the shipped seat briefs exist; findings were not sorted into
regression versus pre-existing; a backlog census was hand-scripted rather than taken from
the tooling; and the full-suite-before-commit rule was broken one commit after being
shipped, leaving main red for six commits.
So: when a rule matters, gate it in the command people actually run - LL0027 in the lessons registry. Adding a line here is the weakest available fix, and a rule stated here with no gate behind it should be read as a known-weak one.
What will refuse you
These block. Everything else in this file is guidance.
| Gate | Refuses |
|---|---|
| pre-commit + commit-msg hooks | any guard failure; a multi-id subject with no Refs: trailer; a collapsed test suite |
sprint plan |
a batch whose units lack Affects: or Points:, or exceed the split threshold |
transition -> Done |
a story whose executable ACs have not passed, or that is past review.two_role_after without both review halves |
transition -> Fixed |
a bug with no parseable Verification depth |
sprint close |
units no independent pass covers; an unanswered checklist item |
critic record |
a verdict carrying no brief provenance (--brief), unless stood down by a recorded config decision |
critic signoff |
a principal the authoring session controls |
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.
- 4d ago First seen · 252 lines · 3,662 tokens per session scan A 57043c81e17c
sdlc-studio AGENTS.md is an instructions file published in the GitHub repository DarrenBenson/sdlc-studio (21 stars, last pushed yesterday), licensed MIT. It adds 3,662 tokens to every session, about $0.0183 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
stockbit-mcp CLAUDE.md
Claude Code instructions for INo-xious/stockbit-mcp, covering claude.md, what this is, commands, the map and three invariants. do not break them; each has a test.
devin-handoff AGENTS.md
Instructions for club-cog/devin-handoff, covering devin handoff — agent guide, what this is, when to hand off, finding the script and how to use.
agent-plugins-skills copilot-instructions.md
Copilot instructions for richfrem/agent-plugins-skills, covering copilot instructions for agent-plugins-skills, 1. think before coding, 2. simplicity first, 3. surgical changes and 4. goal-driven execution.
dev-workspace AGENTS.md
Instructions for iwe-org/dev-workspace, covering agent operating manual, start of every session, the operating loop, conventions and iwe basics.
harmont-cli CLAUDE.md
Instructions for harmont-dev/harmont-cli, covering testing, documentation, dsl and keep the sdk, hm init templates, and docs in sync.
agentic-seo-skills AGENTS.md
Instructions for agencia-conversion/agentic-seo-skills, covering agentic seo — agent runtime contract, mission, six pillars, audience and delivery — web companion first.