sibyl

Solid

Persistent memory and task coordination for this project. Invoke for any prompt about past work, project state, in-progress tasks, prior decisions, gotchas, or capturing a non-obvious learning. Also covers semantic search across project knowledge and external docs.

AI & Automation 59 stars 9 forks Updated today Apache-2.0

Install

View on GitHub

Quality Score: 86/100

Stars 20%
59
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
75
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# Sibyl This stub points at the version-matched workflow shipped by the installed CLI. **Load the compact contract at session start:** ```bash sibyl skill get contract ``` It carries the five agent verbs, small enums, and hard rules in about 1k tokens. Before the first write intent (`remember`, `correct`, or a task mutation), load the full version-matched pack: ```bash sibyl skill get core ``` Load packs whole. Never pipe `sibyl skill get` through `head`/`tail`: the flag contracts and enum tables sit past the fold, and truncated bootstraps are the leading cause of invented-flag errors. Loading a pack is not the memory work itself: after `skill get`, still run the `sibyl context` / task command you came for. Sessions that load the contract and skip the verb defeat the loop. **Quick triggers (intent -> verb):** - "what am i working on" -> `sibyl task list --status doing,blocked` - "where did i leave off" -> `sibyl context "<goal>"` - "have we hit this before" / "do we know X" -> `sibyl context "<topic>"` - "remember this" / "write this up" / "save this insight" -> `sibyl remember "<title>" "<body>"` - "that memory is wrong/stale/superseded" -> `sibyl correct <id>` then add `--action ...` Live status: `sibyl health` shows server reachability. Useful follow-ups: ```bash sibyl skill list sibyl skill get workflows sibyl skill get examples ``` Hooks are separate from skills. Install or update hooks only when the user explicitly wants automatic prompt/session integration...

Details

Author
hyperb1iss
Repository
hyperb1iss/sibyl
Created
8 months ago
Last Updated
today
Language
Python
License
Apache-2.0

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Listed

memory-search

Searches the project's recorded memory (core-memory-kit) — decisions, conventions, architecture, the reasoning behind choices, and where things live — and returns a curated, cited summary. Fire whenever the answer might be something the project already established in past work, HOWEVER the question is phrased — any prior decision, convention, rationale, or "how/where/why is it this way" question, including oblique or roundabout asks ("why is everything so spread out?", "remind me what we settled on for X", "how come these files are tiny?"). Also fire when a "[core-memory-kit] Memory available" hint appears on the prompt. The examples are illustrative, not a checklist — prefer recalling over re-deriving an answer from the code. The session-start snapshot is a bounded index; this skill reaches the rest. Skip only when the question is purely about uncommitted or just-edited live code that memory cannot know, concerns this conversation only, or the user asked to ignore memory.

7 Updated today
LH8PPL
AI & Automation Listed

lythoskill-project-scribe

Session context dump. Self-assess what the conversation contains that has NO other carrier (no task, no ADR, no epic) — pitfalls, working-tree anomalies, why-we-chose-this, specific next steps — and write to daily/YYYY-MM-DD.md. Things WITH structured carriers go to their carriers. Things WITHOUT carriers but needed by the next agent go here. Forms CQRS write-side pair with project-onboarding (read-side).

3 Updated today
lythos-labs
AI & Automation Listed

memarium

Digest already-synced raw_sessions into a per-project typed Memory OS — episodic memories (one per work thread, the arc/dead-ends/decisions), semantic facts, procedural gotchas, plus the derived entity wiki + distilled Q&A. Triggers on `/memarium`. Two modes auto-selected by cwd — project-mode (cwd ≠ session-repo, digests just the matching project) or global-mode (cwd = session-repo, fan-out one subagent per pending project). Per-project isolated. There is ONE knowledge layer — typed memory, read by agents (recall/primer); no human-facing book.

1 Updated 4 days ago
june9593