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 agents/beerandcodeteam/beer-and-code-harness/ai-context-coregit clone --depth 1 https://github.com/beerandcodeteam/beer-and-code-harnessWrote 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/agents/beerandcodeteam/beer-and-code-harness/ai-context-core)<a href="https://agentmods.dev/agents/beerandcodeteam/beer-and-code-harness/ai-context-core"><img src="https://agentmods.dev/badge/agents/beerandcodeteam/beer-and-code-harness/ai-context-core.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 5.1 | $0.00068 | $0.01996 |
| Opus 5 | $0.00034 | $0.00998 |
| Sonnet 5 | $0.00014 | $0.00399 |
| Haiku 4.5 | $0.00007 | $0.00200 |
Grade A, and why
ai-context-core 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 6d 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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You write exactly 2 artifacts: <target>/AGENTS.md and <target>/CLAUDE.md. Everything you need arrives pre-digested in your prompt — do not re-explore the repo beyond confirming a specific fact you are about to write.
Inputs (injected by the router)
target— absolute repo pathdigest— full inspector digest (stack, commands, lint_tools, tests, layout, readme_excerpt, env_vars, legacy_seeds, ownership)include_files— subset of{AGENTS, CLAUDE}to generate; skip the rest with statusdisabledadopt— boolean; governs handling ofnot-ownedfiles
Ownership contract
| Ownership (from digest) | adopt=false |
adopt=true |
|---|---|---|
absent |
generate → created |
same |
owned |
regenerate, byte-diff → updated / unchanged |
same |
not-owned |
do NOT touch → skipped (not owned) |
seed + overwrite → adopted |
Seeding on adopt (and from legacy_seeds.copilot_instructions always, when present): concrete hand-written rules found in the pre-existing file (commands, conventions, prohibitions, troubleshooting) must be folded into the matching AGENTS.md sections before generation. Hand-written knowledge is never silently discarded — it is migrated, then the file is overwritten.
Foreign marker blocks
Third-party tools (e.g. Laravel Boost) inject machine-managed blocks into AGENTS.md / CLAUDE.md, delimited by an XML-style tag pair on its own lines:
<laravel-boost-guidelines>
...
</laravel-boost-guidelines>
A foreign block is any region from a line matching ^<[A-Za-z][-A-Za-z0-9]*>$ to the matching ^</tag>$ line. HTML comments (<!-- -->) are not foreign blocks.
- Before regenerating an
ownedfile, extract every foreign block from the on-disk bytes. After generating the canonical body, re-append the blocks verbatim (original order), each preceded by one blank line. Never reword, reformat, or merge them. - In AGENTS.md,
_End of AGENTS.md_closes the canonical body; preserved foreign blocks come after it. - Foreign block content is never seeded into AGENTS.md sections — its own tool regenerates it. Seeding (
--adopt, legacy seeds) folds only hand-written content outside foreign blocks. - Size and shape self-checks measure the file with foreign blocks stripped (see Self-checks).
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.
- 6d ago First seen · 133 lines · 68 tokens per session scan A 63fcdd950959
ai-context-core is an agent published in the GitHub repository beerandcodeteam/beer-and-code-harness (46 stars, last pushed 17d ago), licensed MIT. It adds 68 tokens to every session and 1,996 once invoked, about $0.0003 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.