Allow customization of system prompt

[draxxris]

I specifically deny certain tools on certain agents, but magic-context is injecting a system prompt about how to use these tools. So then the LLM gets confused and tries to use a tool that doesn't exist. Would be nice if we could rewrite the system prompt entirely for this plugin.

[ualtinok]

Shipped in v0.16.2.

New top-level system_prompt_injection config. skip_signatures is just a list of substrings — if any one of them appears anywhere in an agent's system prompt, that agent gets no ## Magic Context block for that turn (and no project docs / user profile / key files either).

Two ways to use it:

"system_prompt_injection": {
    "enabled": true,                  // set to false to disable injection globally
    "skip_signatures": [
        // Option 1 — match a unique phrase that's already in the agent's prompt:
        "You are a read-only QA agent",

        // Option 2 — drop a marker into your custom agent's prompt and match it here.
        // The exact marker text is up to you; this is just an example convention:
        "<!-- magic-context: skip -->"
    ]
}

Both options skip the same things: the guidance block, project docs, user profile, key files, and the system-prompt hash update (so opted-out calls don't trigger spurious cache busts on later main-agent turns).

Upgrade with:

bunx --bun @cortexkit/magic-context@latest doctor --force

Closing — please reopen if skip_signatures doesn't cover the case you had in mind.