anomalyco / anomalyco/opencode
Feature request: automatic session summarization + skill self-evolution (hermes-style loop)
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 209k
- Forks
- 27.5k
- PR merge metrics
- PR metrics pending
Description
Summary
Add the ability for sessions to be automatically summarized into reusable knowledge, and for skills to self-evolve — a periodic self-review/summary loop that writes distilled learnings back into the knowledge system (similar to hermes-style agents).
Motivation
- Long conversations accumulate reusable learnings that are lost when context gets compacted.
- Skills stay static unless manually edited; they could absorb distilled lessons automatically (self-improvement).
Proposed behavior
- After N turns (or via an explicit command such as
opencode summarize), produce a concise, citable summary of the session's reusable knowledge. - Optionally write that distilled knowledge into an auto-managed knowledge folder (indexed by a local search engine) so future sessions can retrieve it.
- Skills/agents can optionally self-evolve: when a clear recurring pattern is detected, append the experience back into the skill's own definition.
- Rate-limited + opt-in/opt-out (off by default or per-skill toggle) to avoid noise.
Acceptance criteria
- A trigger (command or prompt) that summarizes the current session into a knowledge file.
- The summary is retrievable later (indexed, with source references).
- Optional self-evolution hook for skills, gated behind a flag.
Additional context
Adaptable to existing skill-loading/knowledge conventions in opencode. Happy to iterate on the design in the issue thread.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by tracing the existing skill-loading and knowledge conventions, then locate the command entry point for a possible opencode summarize trigger. Review how the local search index and source references are handled before defining the opt-in, rate-limited session-summary flow and the gated self-evolution hook. Done means summaries are written to retrievable knowledge files and the optional skill update path is covered by the acceptance criteria.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- ai
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100