fix(aider): CONVENTIONS.md is a roster index, not 3.8 million characters of agents (#871)

Aider loads a conventions file into context and keeps it there for the whole
session — that is what the file is for, and the docs say to load it with
--read so prompt caching can hold it. This integration concatenated every
agent body into it. At 279 agents that is 3,816,372 characters, roughly a
million tokens. No model takes that. Anyone who ran

    ./scripts/install.sh --tool aider

got a CONVENTIONS.md that either blows the context window on the first turn
or bills for a million tokens trying.

CONVENTIONS.md is now the roster index it was described as: one entry per
agent with the name, the description, the division, and the path to the
agent file. 96,823 characters, down from 3.8 million. The header explains
how to pull a single agent's full instructions into the session:

    /read-only /path/to/agency-agents/engineering/engineering-frontend-developer.md

Naming an agent in a prompt still works the way it did — the description is
what the model needed for that, and it is still there.

test-convert-outputs.sh now holds the index to being an index: it fails if
CONVENTIONS.md grows past 250,000 characters, if it does not list exactly
one path per roster agent, or if any path it prints does not resolve. The
existing round-trip check on the accumulated file still covers the names and
descriptions.

README.md

@@ -733,7 +733,7 @@ The Agency works natively with Claude Code, and ships conversion + install scrip
 - **[Gemini CLI](https://github.com/google-gemini/gemini-cli)** -- `.md` agent files -> `~/.gemini/agents/`
 - **[OpenCode](https://opencode.ai)** — `.md` agent files → `.opencode/agents/`
 - **[Cursor](https://cursor.sh)** — `.mdc` rule files → `.cursor/rules/`
-- **[Aider](https://aider.chat)** — single `CONVENTIONS.md` → `./CONVENTIONS.md`
+- **[Aider](https://aider.chat)** — `CONVENTIONS.md` roster index → `./CONVENTIONS.md`
 - **[Windsurf](https://codeium.com/windsurf)** — single `.windsurfrules` → `./.windsurfrules`
 - **[OpenClaw](https://github.com/openclaw/openclaw)** — `SOUL.md` + `AGENTS.md` + `IDENTITY.md` per agent
 - **[Qwen Code](https://github.com/QwenLM/qwen-code)** — `.md` SubAgent files → `~/.qwen/agents/`
@@ -926,7 +926,10 @@ See [integrations/cursor/README.md](integrations/cursor/README.md) for details.
 <details>
 <summary><strong>Aider</strong></summary>
 
-All agents are compiled into a single `CONVENTIONS.md` file that Aider reads automatically.
+`CONVENTIONS.md` is the roster index — every agent's name, description, and the
+path to its full instructions. Aider keeps a conventions file in context for the
+whole session, and the 279 bodies together are about a million tokens, so the
+file lists the agents rather than inlining them.
 
 ```bash
 cd /your/project
@@ -938,6 +941,11 @@ Then reference agents in your Aider session:
 Use the Frontend Developer agent to refactor this component.
 ```
 
+When you want an agent's full instructions, read its file in:
+```
+/read-only /path/to/agency-agents/engineering/engineering-frontend-developer.md
+```
+
 See [integrations/aider/README.md](integrations/aider/README.md) for details.
 </details>