Rebase of PR #1922 onto current dev: the ~14 files of v6.1.0-era codex/release drift are dropped, the porting-guide edits (stale against the post-prune rewrite, no Hermes content) are dropped, and the Hermes surface is kept intact: .hermes-plugin/ (on_session_start bootstrap injection), tests/hermes/ (20 tests, passing), docs/README.hermes.md, references/hermes-tools.md, the Platform Adaptation row, README section, and Python ignores. Known open items from review, unchanged by this rebase: the injection mechanism uses ctx.inject_message from on_session_start, which the official plugin guide does not document (pre_llm_call returning {"context": ...} is the sanctioned path), skills are not registered via ctx.register_skill, and the acceptance transcript predates the fix. Co-authored-by: kumarabd <kumarabd@users.noreply.github.com>
919 B
Hermes Agent
Superpowers supports Hermes Agent via an in-process Python plugin (Shape B).
Install
hermes plugins install obra/superpowers --enable
What you get
All Superpowers skills auto-trigger in Hermes sessions:
brainstorming before feature work, systematic-debugging on bugs,
test-driven-development for implementation, writing-plans before
touching code, and all other skills in skills/.
How it works
The plugin registers an on_session_start hook with the Hermes plugin API.
At the start of each session, the hook injects the using-superpowers bootstrap
as a user-role message via ctx.inject_message(role="user"). A session-id guard
prevents double-injection if the hook fires more than once per session.
Skills are loaded on demand during the session using skill_view("skill-name").
Verifying
See .hermes-plugin/INSTALL.md for the smoke check and acceptance test.